TECHNICAL FIELD
[0001] The present disclosure relates to a multi-junction (also called a tandem-type, a
stack-type, or a lamination-type) solar cell, a photoelectric conversion device, and
a compound-semiconductor-layer lamination structure that use a compound semiconductor.
BACKGROUND ART
[0002] As a compound semiconductor configured of two or more types of elements, many types
exist depending on a combination of the elements. Also, by laminating a lot of compound
semiconductor layers made of different materials, a compound semiconductor device
having various functions and various physical properties are achievable. As an example
thereof, a solar cell may be mentioned. Here, as a solar cell, a silicon-based solar
cell that uses silicon as a semiconductor, a compound semiconductor solar cell that
uses a compound semiconductor, an organic solar cell that uses an organic material,
etc. may be mentioned. In particular, the compound semiconductor solar cell has been
developed aiming further improvement in energy conversion efficiency.
[0003] As a means for improving energy conversion efficiency of the compound semiconductor
solar cell, there are provided a method in which a plurality of sub-cells each configured
of a thin-film solar cell that is configured of a plurality of compound semiconductor
layers are laminated to form a multi-junction solar cell, a method in which an effective
combination of compound semiconductor materials configuring the compound semiconductor
layers are searched, etc. Each of compound semiconductors such as GaAs and InP has
a unique band gap, and a wavelength of light to be absorbed is different depending
on this difference in band gap. Therefore, by laminating a plurality of types of sub-cells,
efficiency of absorption of solar light that has a wide wavelength range is improved.
In lamination, a combination of lattice constants and physical property values (such
as band gaps) of crystal structures of the compound semiconductors configuring the
respective sub-cells is important.
[0004] By the way, most of the multi-junction solar cells under current consideration are
classified into a lattice-matched type and a lattice-mismatched type. In the lattice-matched
type, compound semiconductor layers are laminated that are made of compound semiconductors
having lattice constants that are almost the same with one another. In the lattice-mismatched
type, compound semiconductor layers are laminated that are made of compound semiconductors
having lattice constants that are different from one another with the use of metamorphic
growth accompanied by dislocation. However, in the metamorphic growth method, undesirable
lattice mismatch inevitably occurs, and therefore, there is an issue that quality
of the compound semiconductors is significantly lowered.
[0005] On the other hand, in recent years, there has been proposed a method of manufacturing
a multi-junction solar cell that utilizes a substrate bonding technique in junction
of compound semiconductor layers, and a four-junction solar cell that has a structure
of In
0.48Ga
0.52P/GaAs/InGaAsP/In
0.53Ga
0.47As has been reported.
[0006] This substrate bonding technique is a technique to form homojunction or heterojunction
between the compound semiconductor layers to be joined, and may be classified, for
example, into a direct bonding scheme in which different compound semiconductor layers
are bonded directly to one another (for example, see Non-patent Literature 1: "
Wafer Bonding and Layer Transfer Processes for High Efficiency Solar Cells", NCPV
and Solar Program Review Meeting 2003), and a scheme in which the compound semiconductor layers are joined with a connection
layer in between. The substrate bonding technique has an advantage that it is not
accompanied by an increase in threading dislocation. Existence of the threading dislocation
leads to a not-preferable effect on electron performance of the compound semiconductor
layers. In particular, the existence of the threading dislocation provides an easy
diffusion path in the compound semiconductor layers as with a dopant and a recombination
center, and causes a decrease in carrier density of the compound semiconductor layers.
Also, the substrate bonding technique resolves the issue of lattice mismatch, and
further avoids epitaxial growth caused by the lattice mismatch. Therefore, threading
dislocation density that degrades the performance of the solar cell is largely reduced.
In this substrate bonding technique, a covalent bonding is formed in an interface
between different substances, in particular, in a hetero interface. At this time,
it is important to perform a substrate junction process at a temperature by which
thermal variation does not exceed a dynamic barrier necessary for progression in threading
dislocation.
[0007] In junction by the direct bonding scheme, semiconductor-semiconductor bonding is
performed in a nuclear scale. Therefore, transparency, heat conductivity, heat resistance,
and reliability of the junction portion are superior than those in a case where junction
is formed with the use of metal paste, a glass raw material (frit), etc. In this direct
bonding scheme, an integrated-type or two-terminal compound semiconductor device is
allowed to be integrated to a module with simplicity equivalent to that in a solar
cell configured of a single-junction device, specifically, only by alloying the respective
compound semiconductor layers to be laminated.
CITATION LIST
NON-PATENT LITERATURE
[0008] Non-patent Literature 1: Wafer Bonding and Layer Transfer Processes for High Efficiency
Solar Cells, NCPV and Solar Program Review Meeting 2003
SUMMARY OF THE INVENTION
[0009] By the way, in order to improve efficiency in utilizing solar light, it is necessary
to take in solar light spectrum in a wide range. A maximum wavelength of the solar
light spectrum is 2.5 µm. On the other hand, for example, in the above-described Non-patent
Literature 1, the lowermost layer is configured of an InGaAs layer having a band gap
of 0.72 eV, and this InGaAs layer is capable of only taking in the solar light having
a wavelength of about 1.7 µm. In order to take in the solar light spectrum in a wide
range, it is necessary to configure the sub-cell with the use of a compound semiconductor
layer that has a further-lower band gap value. However, as long as the present inventors
have examined, a technology is not known to provide a sub-cell formed of a compound
semiconductor that has a desirable band gap while allowing, as a whole, a lattice
constant of a base (for example, a substrate for film formation) for forming such
a sub-cell to be matched with a lattice constant of the compound semiconductor that
configures the sub-cell.
[0010] Therefore, it is desirable to provide a multi-junction solar cell in which lattice
match is established, as a whole, with a material configuring the base and that includes
a sub-cell formed of a compound semiconductor having a desirable band gap, or a photoelectric
conversion device or a compound-semiconductor-layer lamination structure that includes
a compound semiconductor layer.
[0011] A multi-junction solar cell according to an embodiment of the present disclosure
includes a plurality of sub-cells that are laminated, in which light enters from the
sub-cell located in an uppermost layer to the sub-cell located in a lowermost layer,
and electric power is generated in each of the sub-cells,
each of the sub-cells includes a first compound semiconductor layer and a second compound
semiconductor layer that are laminated, the first compound semiconductor layer having
a first conductivity type, and the second compound semiconductor layer having a second
conductivity type that is different from the first conductivity type,
in at least one predetermined sub-cell of the plurality of sub-cells,
the first compound semiconductor layer is configured of at least one first-compound-semiconductor-layer
lamination unit in which a 1-A compound semiconductor layer and a 1-B compound semiconductor
layer are laminated,
the second compound semiconductor layer is configured of at least one second-compound-semiconductor-layer
lamination unit in which a 2-A compound semiconductor layer and a 2-B compound semiconductor
layer are laminated,
a compound semiconductor composition configuring the 1-A compound semiconductor layer
and a compound semiconductor composition configuring the 2-A compound semiconductor
layer are same compound semiconductor composition A,
a compound semiconductor composition configuring the 1-B compound semiconductor layer
and a compound semiconductor composition configuring the 2-B compound semiconductor
layer are same compound semiconductor composition B,
the compound semiconductor composition A is determined based on a value of a band
gap of the predetermined sub-cell,
the compound semiconductor composition B is determined based on a difference between
a base lattice constant of a base used at a time of forming the first and second compound
semiconductor layers and a lattice constant of the compound semiconductor composition
A,
a thickness of the 1-B compound semiconductor layer is determined based on a difference
between the base lattice constant and a lattice constant of the compound semiconductor
composition B, and on a thickness of the 1-A compound semiconductor layer,
a thickness of the 2-B compound semiconductor layer is determined based on the difference
between the base lattice constant and the lattice constant of the compound semiconductor
composition B, and on a thickness of the 2-A compound semiconductor layer,
the thicknesses of the 1-A compound semiconductor layer and the 2-A compound semiconductor
layer are smaller than a critical film thickness of the compound semiconductor composition
A, and are thicknesses that cause no quantum effect, and
the thicknesses of the 1-B compound semiconductor layer and the 2-B compound semiconductor
layer are smaller than a critical film thickness of the compound semiconductor composition
B, and are thicknesses that cause no quantum effect.
[0012] A photoelectric conversion device according to an embodiment of the present disclosure
includes a first compound semiconductor layer and a second compound semiconductor
layer that are laminated, the first compound semiconductor layer having a first conductivity
type, and the second compound semiconductor layer having a second conductivity type
that is different from the first conductivity type,
the first compound semiconductor layer is configured of at least one first-compound-semiconductor-layer
lamination unit in which a 1-A compound semiconductor layer and a 1-B compound semiconductor
layer are laminated,
the second compound semiconductor layer is configured of at least one second-compound-semiconductor-layer
lamination unit in which a 2-A compound semiconductor layer and a 2-B compound semiconductor
layer are laminated,
a compound semiconductor composition configuring the 1-A compound semiconductor layer
and a compound semiconductor composition configuring the 2-A compound semiconductor
layer are same compound semiconductor composition A,
a compound semiconductor composition configuring the 1-B compound semiconductor layer
and a compound semiconductor composition configuring the 2-B compound semiconductor
layer are same compound semiconductor composition B,
the compound semiconductor composition A is determined based on a value of a band
gap of the photoelectric conversion device,
the compound semiconductor composition B is determined based on a difference between
a base lattice constant of a base used at a time of forming the first and second compound
semiconductor layers and a lattice constant of the compound semiconductor composition
A,
a thickness of the 1-B compound semiconductor layer is determined based on a difference
between the base lattice constant and a lattice constant of the compound semiconductor
composition B, and on a thickness of the 1-A compound semiconductor layer,
a thickness of the 2-B compound semiconductor layer is determined based on the difference
between the base lattice constant and the lattice constant of the compound semiconductor
composition B, and on a thickness of the 2-A compound semiconductor layer,
the thicknesses of the 1-A compound semiconductor layer and the 2-A compound semiconductor
layer are smaller than a critical film thickness of the compound semiconductor composition
A, and are thicknesses that cause no quantum effect, and
the thicknesses of the 1-B compound semiconductor layer and the 2-B compound semiconductor
layer are smaller than a critical film thickness of the compound semiconductor composition
B, and are thicknesses that cause no quantum effect.
[0013] A compound-semiconductor-layer lamination structure according to an embodiment of
the present disclosure includes at least one compound-semiconductor-layer lamination
unit in which an A compound semiconductor layer and a B compound semiconductor layer
are laminated,
a compound semiconductor composition B configuring the B compound semiconductor layer
is determined based on a difference between a base lattice constant of a base used
at a time of forming the A and B compound semiconductor layers and a lattice constant
of a compound semiconductor composition A configuring the A compound semiconductor
layer,
a thickness of the B compound semiconductor layer is determined based on a difference
between the base lattice constant and a lattice constant of the compound semiconductor
composition B, and on a thickness of the A compound semiconductor layer,
the thickness of the A compound semiconductor layer is smaller than a critical film
thickness of the compound semiconductor composition A, and is a thickness that causes
no quantum effect, and
the thicknesses of the B compound semiconductor layer is smaller than a critical film
thickness of the compound semiconductor composition B, and is a thickness that causes
no quantum effect.
[0014] In the predetermined sub-cell in the multi-junction solar cell of an embodiment of
the present disclosure, or in the photoelectric conversion device: the first compound
semiconductor layer and the second compound semiconductor layer are laminated; the
first compound semiconductor layer is configured of at least one first-compound-semiconductor-layer
lamination unit in which the 1-A compound semiconductor layer and the 1-B compound
semiconductor layer are laminated; and the second compound semiconductor layer is
configured of at least one second-compound-semiconductor-layer lamination unit in
which the 2-A compound semiconductor layer and the 2-B compound semiconductor layer
are laminated. Further, the compound semiconductor composition A that configures the
1-A compound semiconductor layer and the 2-A compound semiconductor layer is determined
based on the value of a band gap of the predetermined sub-cell; the compound semiconductor
composition B that configures the 1-B compound semiconductor layer and the 2-B compound
semiconductor layer is determined based on the difference between the base lattice
constant of the base used at the time of forming the first and second compound semiconductor
layers and the lattice constant of the compound semiconductor composition A; and the
thicknesses of the 1-B compound semiconductor layer and the 2-B compound semiconductor
layer are determined based on the difference between the base lattice constant and
the lattice constant of the compound semiconductor composition B, and on the thickness
of the 1-A compound semiconductor layer and the thickness of the 2-A compound semiconductor
layer. Moreover, the thicknesses of the 1-A compound semiconductor layer and the 2-A
compound semiconductor layer are smaller than the critical film thickness of the compound
semiconductor composition A, and are thicknesses that cause no quantum effect; and
the thicknesses of the 1-B compound semiconductor layer and the 2-B compound semiconductor
layer are smaller than the critical film thickness of the compound semiconductor composition
B, and are thicknesses that cause no quantum effect. In such a manner, a distortion
compensation structure is configured of the 1-A compound semiconductor layer and the
1-B compound semiconductor layer.
[0015] Further, when a desirable band gap of the predetermined sub-cell, in other words,
for example, a wavelength of light which the predetermined sub-cell is allowed to
absorb most efficiently is set, or when a desirable band gap of the photoelectric
conversion device, in other words, a wavelength of light which the photoelectric conversion
device is allowed to absorb most efficiently or a desirable light emission wavelength
is set, the compound semiconductor composition A that achieves this is determined.
However, usually, the lattice constant of the determined compound semiconductor composition
A and the base lattice constant have a difference, in other words, often, lattice-unmatched
type is established. Therefore, in order to eliminate this difference, in other words,
in order to cancel this difference (in order to establish a lattice-matched type),
the compound semiconductor composition B is determined. Moreover, based on the difference
between the base lattice constant and the lattice constant of the compound semiconductor
composition B, and on the thicknesses of the 1-A compound semiconductor layer and
the 2-A compound semiconductor layer, not only the thicknesses of the 1-B compound
semiconductor layer and the 2-B compound semiconductor layer are determined, but also,
upper-limit values and lower-limit values of layer thicknesses of the 1-A and 2-A
compound semiconductor layers and the 1-B and 2-B compound semiconductor layers are
defined. In other words, when these layer thicknesses are over the upper-limit values,
lattice mismatch is caused, and dislocation may be caused. Further, when these layer
thicknesses are smaller than the lower-limit values, the value of the band gap is
varied.
[0016] In such a manner, in each of the first compound semiconductor layer and the second
compound semiconductor layer in the predetermined sub-cell, the compound semiconductor
composition is optimized (in other words, the band gap and the lattice constant are
optimized), and the layer thickness is optimized. As a result of the above, even when
lattice mismatch is established between the 1-A compound semiconductor layer and the
base, this lattice mismatch is cancelled by the 1-B compound semiconductor layer,
and the first compound semiconductor layer as a whole becomes of a lattice-matched
type. The same is applicable also to the second compound semiconductor layer as a
whole. Moreover, the first compound semiconductor layer and the second compound semiconductor
layer as a whole are allowed to achieve efficient absorption with a desirable wavelength
of light or light emission with a desirable wavelength. Further, this may, for example,
further improve efficiency in absorption of solar light having a wide wavelength range,
in the multi-junction solar cell.
[0017] In the compound-semiconductor-layer lamination structure according to an embodiment
of the present disclosure: the compound semiconductor composition B is determined
based on the difference between the base lattice constant and the lattice constant
of the compound semiconductor composition A; and the thickness of the B compound semiconductor
layer is determined based on the difference between the base lattice constant and
the lattice constant of the compound semiconductor composition B, and on the thickness
of the A compound semiconductor layer. Therefore, even when lattice mismatch is established
between the A compound semiconductor layer and the base, this lattice mismatch is
cancelled by the B compound semiconductor layer, and the compound-semiconductor-layer
lamination structure as a whole becomes of a lattice-matched type. Therefore, restriction
such as that it is necessary to determine the compound semiconductor composition A
being limited by the base lattice constant in order to obtain the lattice-matched
type is eased. Therefore, it is possible to expand the range of options for the compound
semiconductor composition of the compound semiconductor layer configuring the compound-semiconductor-layer
lamination structure, and to improve freedom in options.
BRIEF DESCRIPTION OF THE DRAWINGS
[0018]
[FIG. 1] (A) and (B) of FIG. 1 are a conceptual diagram of a multi-junction solar
cell, and a conceptual diagram of a predetermined sub-cell located in a lowermost
layer, a photoelectric conversion device, and a compound-semiconductor-layer lamination
structure, respectively, in Example 1 or Example 3.
[FIG. 2] (A) and (B) of FIG. 2 are conceptual diagrams of a compound semiconductor
layer and the like for explaining a method of manufacturing the multi-junction solar
cell, the photoelectric conversion device, and the compound-semiconductor-layer lamination
structure in Example 3.
[FIG. 3] (A) and (B) of FIG. 3 are conceptual diagrams of the compound semiconductor
layer and the like for explaining the method of manufacturing the multi-junction solar
cell, the photoelectric conversion device, and the compound-semiconductor-layer lamination
structure in Example 3, following (B) of FIG. 2.
[FIG. 4] (A) and (B) of FIG. 4 are conceptual diagrams of multi-junction solar cells,
photoelectric conversion devices, and compound-semiconductor-layer lamination structures
in Example 4 and Example 5, respectively.
[FIG. 5] (A) and (B) of FIG. 5 are conceptual diagrams of a compound semiconductor
layer and the like for explaining a method of manufacturing a multi-junction solar
cell, a photoelectric conversion device, and a compound-semiconductor-layer lamination
structure in Example 6.
[FIG. 6] (A) and (B) of FIG. 6 are conceptual diagrams of the compound semiconductor
layer and the like for explaining the method of manufacturing the multi-junction solar
cell, the photoelectric conversion device, and the compound-semiconductor-layer lamination
structure in Example 6, following (B) of FIG. 5.
[FIG. 7] FIG. 7 is a conceptual diagram of the compound semiconductor layer and the
like for explaining the method of manufacturing the multi-junction solar cell, the
photoelectric conversion device, and the compound-semiconductor-layer lamination structure
in Example 6, following (B) of FIG. 6.
[FIG. 8] (A) and (B) of FIG. 8 are conceptual diagrams of a compound semiconductor
layer and the like for explaining a method of manufacturing a multi-junction solar
cell, a photoelectric conversion device, and a compound-semiconductor-layer lamination
structure in Example 7.
[FIG. 9] (A) and (B) of FIG. 9 are conceptual diagrams of the compound semiconductor
layer and the like for explaining the method of manufacturing the multi-junction solar
cell, the photoelectric conversion device, and the compound-semiconductor-layer lamination
structure in Example 7, following (B) of FIG. 8.
[FIG. 10] (A) and (B) of FIG. 10 are conceptual diagrams of the compound semiconductor
layer and the like for explaining the method of manufacturing the multi-junction solar
cell, the photoelectric conversion device, and the compound-semiconductor-layer lamination
structure in Example 7, following (B) of FIG. 9.
[FIG. 11] FIG. 11 is a conceptual diagram of a modification of the multi-junction
solar cell in Example 1.
[FIG. 12] FIG. 12 is a graph showing results obtained by variously varying band gaps
of a second sub-cell and a first sub-cell, simulating, and determining conversion
efficiencies, when a band gap of a fourth sub-cell is 1.910 eV and a band gap of a
third sub-cell is 1.420 eV in a multi-junction solar cell in which four sub-cells
are laminated.
[FIG. 13] FIG. 13 is a characteristic diagram illustrating film-forming characteristics
of metal atoms.
[FIG. 14] FIG. 14 is a characteristic diagram illustrating a relationship between
a film thickness of a Ti layer and light transmittance.
[FIG. 15] (A) and (B) of FIG. 15 are photographs showing a result of an infrared microscopic
transmission experiment.
[FIG. 16] FIG. 16 is a graph showing a relationship between photon energy and an absorption
coefficient for each concentration of a p-type dopant in a p-type GaAs layer.
[FIG. 17] FIG. 17 is a graph showing a relationship between a thickness of the p-type
GaAs layer at p-type dopant concentration of 3×1019 and transmittance of solar light at maximum wavelength of 2.5 µm.
[FIG. 18] FIG. 18 is a photograph of a bright-field image, of an interface of a junction
of an InP substrate and a GaAs substrate, obtained by a scanning transmission electron
microscope.
[FIG. 19] FIG. 19 is a graph showing variation in the film thickness of the Ti layer
and the light transmittance over time.
[FIG. 20] FIG. 20 is a graph showing variation in the film thickness of the Ti layer
and the light transmittance over time.
[FIG. 21] FIG. 21 is a graph showing a result of quantitative analysis of concentration
of each atom in each distance in a lamination direction of the multi-junction solar
cell in Example 3 based on energy dispersive X-ray spectrometry.
[FIG. 22] FIG. 22 is a photograph of a cross-section of a bonding junction interface
obtained by a transmission electron microscope.
MODES FOR CARRYING OUT THE INVENTION
[0019] The present disclosure will be described below based on Examples referring to the
drawings. However, the present disclosure is not limited to the Examples, and various
numerical values, materials, etc. in the Examples are examples. It is to be noted
that the description will be given in the following order.
- 1. Description related to general matters of a multi-junction solar cell, a photoelectric
conversion device, and a compound-semiconductor-layer lamination structure of the
present disclosure
- 2. Example 1 (the multi-junction solar cell, the photoelectric conversion device,
and the compound-semiconductor-layer lamination structure of the present disclosure)
- 3. Example 2 (a modification of Example 1)
- 4. Example 3 (another modification of Example 1)
- 5. Example 4 (a modification of Example 3)
- 6. Example 5 (another modification of Example 3)
- 7. Example 6 (still another modification of Example 3)
- 8. Example 7 (a modification of Example 6) and others
[0020] [Description related to general matters of a multi-junction solar cell, a photoelectric
conversion device, and a compound-semiconductor-layer lamination structure of the
present disclosure]
[0021] In a multi-junction solar cell of the present disclosure, a form may be adopted in
which a predetermined sub-cell is located in a lowermost layer. A lamination order
in a plurality of sub-cells is set to be a lamination order by which a band gap of
a compound semiconductor configuring the sub-cell becomes larger as the compound semiconductor
is closer to a light incident side, that is, a lamination order by which the band
gaps of the compound semiconductors are larger in order from a support substrate side
to a second electrode side which will be described later. In some cases, part of the
plurality of sub-cells may be configured of a Ge layer.
[0022] In the multi-junction solar cell of the present disclosure including the above-described
preferable form, or the photoelectric conversion device of the present disclosure
or the compound-semiconductor-layer lamination structure of the present disclosure,
a form may be adopted in which a group of atoms configuring the compound semiconductor
composition A is same as a group of atoms configuring the compound semiconductor composition
B. Further, in this case, a form may be adopted in which atomic percentage of the
group of atoms configuring the compound semiconductor composition A is different from
atomic percentage of the group of atoms configuring the compound semiconductor composition
B.
[0023] Moreover, a value of the band gap of the predetermined sub-cell in the multi-junction
solar cell of the present disclosure including the above-described various preferable
forms, or a value of the band gap of the photoelectric conversion device in the photoelectric
conversion device of the present disclosure including the above-described various
preferable forms may be preferably from 0.45 eV to 0.75 eV both inclusive.
[0024] Further, in the multi-junction solar cell of the present disclosure including the
above-described preferable forms and configurations, or in the photoelectric conversion
device of the present disclosure or the compound-semiconductor-layer lamination structure
of the present disclosure, a value of a band gap of the compound semiconductor composition
B may be desirably larger than a value of a band gap of the compound semiconductor
composition A.
[0025] Further, in the multi-junction solar cell or the photoelectric conversion device
of the present disclosure including the above-described preferable forms and configurations,
or in the compound-semiconductor-layer lamination structure of the present disclosure,

and

may be preferably satisfied where LC
A is the lattice constant of the compound semiconductor composition A, LC
B is the lattice constant of the compound semiconductor composition B, and LC
0 is the base lattice constant. Further, in this case,

and

may be more preferably established.
[0026] Further, in the multi-junction solar cell or the photoelectric conversion device
of the present disclosure including the above-described preferable forms and configurations,
or in the compound-semiconductor-layer lamination structure of the present disclosure,

may be desirably satisfied where LC
A is the lattice constant of the compound semiconductor composition A, LC
B is the lattice constant of the compound semiconductor composition B, LC
0 is the base lattice constant, t
A is the thickness of the 1-A compound semiconductor layer and the 2-A compound semiconductor
layer (or the thickness of the A compound semiconductor layer), and t
B is the thickness of the 1-B compound semiconductor layer and the 2-B compound semiconductor
layer (or the thickness of the B compound semiconductor layer).
[0027] Further, in the multi-junction solar cell or the photoelectric conversion device
of the present disclosure including the above-described various preferable forms and
configurations, or in the compound-semiconductor-layer lamination structure of the
present disclosure, a configuration may be adopted in which the base is formed of
InP, the compound semiconductor composition A is In
xGa
1-xAs, and the compound semiconductor composition B is In
yGa
1-yAs (where x>y). Further, in this case,

and

may be more preferably established. Alternatively, a configuration may be adopted
in which the base is formed of InP, the compound semiconductor composition A is (InP)
1-z(In
x'Ga
1-x'As)
z, and the compound semiconductor composition B is (InP)
1-z(In
y'Ga
1-y'As)
z (where x'>y'). Further, in this case,

and

may be more preferably established.
[0028] In the multi-junction solar cell or the photoelectric conversion device of the present
disclosure including the above-described various preferable forms and configurations,
or in the compound-semiconductor-layer lamination structure of the present disclosure
(hereinafter, they may be collectively and simply referred to as "present disclosure"
in some cases), the thicknesses of the 1-A compound semiconductor layer and the 2-A
compound semiconductor layer are smaller than a critical film thickness of the compound
semiconductor composition A, the thicknesses of the 1-B compound semiconductor layer
and the 2-B compound semiconductor layer are smaller than a critical film thickness
of the compound semiconductor composition B, the thickness of the A compound semiconductor
layer is smaller than a critical film thickness of the compound semiconductor composition
A, and the thicknesses of the B compound semiconductor layer is smaller than a critical
film thickness of the compound semiconductor composition B. Here, when the thickness
of the compound semiconductor layer is smaller than the critical film thickness, lattice
mismatch is reduced due to distortion of crystalline lattice of the compound semiconductor
crystal configuring the compound semiconductor layer, and thereby, occurrence of dislocation
is suppressed. Further, the thicknesses of the 1-A compound semiconductor layer and
the 2-A compound semiconductor layer are thicknesses that cause no quantum effect,
the thicknesses of the 1-B compound semiconductor layer and the 2-B compound semiconductor
layer are thicknesses that cause no quantum effect, the thickness of the A compound
semiconductor layer is a thickness that causes no quantum effect, and the thickness
of the B compound semiconductor layer is a thickness that causes no quantum effect.
Here, "cause no quantum effect" means that quantum level is not generated even when
surrounded by a double hetero barrier when the film thickness is equal to or larger
than the de Broglie wave (from several nanometers to 20 nm both inclusive).
[0029] When the plurality of sub-cells are laminated in the multi-junction solar cell of
the present disclosure, the lattice-mismatch type may be established between the sub-cells
in some cases depending on the lattice constants of the compound semiconductors of
the compound semiconductor layers configuring the sub-cells. In such a case, the sub-cell
and the sub-cell that are to be the lattice-mismatched type may be desirably connected
(joined) with the use of an amorphous connection layer formed of an electrically-conductive
material. Alternatively, also in a case where the photoelectric conversion device
of the present disclosure and a photoelectric conversion device other than that of
the present disclosure are laminated, the lattice-mismatched type may be established
in some cases depending on lattice constants of the compound semiconductors configuring
the adjacent compound semiconductor layers between the laminated photoelectric conversion
devices. In such a case, the compound semiconductor layer and the compound semiconductor
layer that are to be the lattice-mismatched type may be desirably connected (joined)
with the use of an amorphous connection layer formed of an electrically-conductive
material. By connecting (joining) with the use of the amorphous connection layer formed
of the electrically-conductive material in such a manner, contact resistance of a
junction interface of the sub-cells or the compound semiconductor layers is reduced,
and energy conversion efficiency is improved. It is to be noted that, in the description
below, the multi-junction solar cell or the laminated photoelectric conversion devices
provided with such a connection layer may be collectively referred to as "multi-junction
solar cell of the present disclosure and the like provided with a connection layer"
in some cases.
[0030] In the multi-junction solar cell of the present disclosure and the like provided
with the connection layer in such a manner, depending on the lattice constants of
the compound semiconductors that configure adjacent sub-cells or adjacent compound
semiconductor layers, the adjacent sub-cells or the adjacent compound semiconductor
layers may be of a lattice-matched type or a lattice-mismatched type. However, in
the multi-junction solar cell of the present disclosure and the like provided with
the connection layer as a whole, these lattice-matched type/ lattice-mismatched type
exist in a mixed manner. In the multi-junction solar cell of the present disclosure
and the like provided with the connection layer, where a lattice constant of a compound
semiconductor that configures one sub-cell or one compound semiconductor layer that
is adjacent to the connection layer is Lc
1, and a lattice constant of a compound semiconductor that configures the other sub-cell
or the other compound semiconductor layer that is adjacent to the connection layer
is Lc
2, being the lattice-mismatched type means, being a type that satisfies:

or

[0031] It is to be noted that, when a value of (Lc
1-Lc
2)/Lc
1 is out of the above-described range, that is, when the value satisfies:

it is the lattice-matched type. Further, in the multi-junction solar cell of the present
disclosure and the like provided with the connection layer including such a preferable
form, it may be preferable to adopt a form in which a tunnel junction layer is provided
in a place between the adjacent sub-cells in which the connection layer is not provided.
The same is applicable to, or the same may be adopted in the multi-junction solar
cell and the photoelectric conversion device of the present disclosure including the
above-described various preferable forms and configurations.
[0032] Here, as described above, the connection layer is a layer having amorphous characteristics,
and is configured of metal or alloy. Specifically, as a material configuring the connection
layer, it may be preferable to use a material having ohmic characteristics with respect
to the compound semiconductor layer to be connected, more specifically, metal or alloy
having a work function that is smaller than a Fermi level of an n-type semiconductor
or is larger than a Fermi level of a p-type semiconductor. Accordingly, contact resistance
is significantly reduced, and favorable ohmic connection is achievable. It is to be
noted that "amorphous connection layer" or "connection layer having amorphous characteristics"
means a state that does not have long-term orderly characteristics as in crystal and
that does not allow a lattice image to be observed in an image taken by a transmission
electron microscope as shown in FIG. 22.
[0033] A metal thin film (having a thickness of several nanometers or smaller, for example)
is typically formed based on a PVD method such as a vacuum evaporation method and
a sputtering method. However, at that time, the metal thin film is often formed in
an island-like shape and is rarely formed in a layer-like shape. Further, when the
metal thin film is formed in an island-like shape, it is difficult to control the
film thickness with high accuracy. In formation of a metal thin film based on the
vacuum evaporation method, often, an island is formed after atoms and molecules absorbed
on a base have undergone processes such as surface diffusion, collision and aggregation,
and desorption, and the island grows to be combined with an adj acent island, thereby
forming a continuous thin film. At that time, formation of an island, transition from
being amorphous to being a crystal layer, variation in crystal alignment, etc. occur.
[0034] Moreover, when evaporation is performed in a case where the thickness of the metal
thin film is set to about that of a monoatomic layer, it is considered that whether
metal atoms have a two-dimensional layer structure or a three-dimensional island structure
depends on interaction of binding energy between metal atoms in an uppermost face
and metal atoms existing therebelow, and binding energy between the metal atoms and
the base. When the metal atoms are more stable to be combined to the metal atoms,
the metal atoms have the three-dimensional island structure. On the other hand, when
the metal atoms are more stable to be combined with the base, the metal atoms have
the two-dimensional layer structure. FIG. 13 illustrates analogized characteristics
of binding energy between metal atoms and metal atoms for the respective metal atoms
[binding energy between adatoms] and analogized characteristics of binding energy
between metal atoms and the base (here, GaAs or InP) [binding energy between adatom
and substrate]. Metal atoms in Group (A) and Group (B) located in an upper region
of a dashed line in FIG. 13 have high binding energy with respect to the base, and
therefore, it is considered that the metal atoms are allowed to have the two-dimensional
layer structure. Accordingly, as the material of the connection layer, it may be preferable
to use the metal atoms belonging to Group (A) or Group (B).
[0035] Therefore, it may be preferable to adopt a form in which the connection layer made
of an electrically-conductive material, more specifically, the connection layer configured
of metal or alloy includes at least one type of atom (metal atom) selected from a
group consisting of titanium (Ti), aluminum (Al), zirconium (Zr), hafnium (Hf), tungsten
(W), tantalum (Ta), molybdenum (Mo), niobium (Nb), and vanadium (V). It is to be noted
that, also when atom such as iron (Fe), chromium (Cr), nickel (Ni), and aluminum (Al)
is further included in the connection layer, the characteristics thereof is not influenced
at all. Alternatively, it may be preferable to adopt a form in which the connection
layer is configured of a material selected from a group consisting of aluminum-oxide-doped
zinc oxide [AZO], indium-zinc composite oxide [IZO], gallium-doped zinc oxide [GZO],
indium-gallium composite oxide [IGO], In-GaZnO
4 [IGZO], and indium-tin composite oxide [ITO], that is, a material that is transparent
and has electric conductivity. Alternatively, it is possible to adopt a form in which
the connection layer is made of an amorphous compound semiconductor, specifically,
portion (to be noted, this is amorphous) of a compound semiconductor layer in an interface
of a compound semiconductor layer and a compound semiconductor layer. In such a form,
the issue of lattice mismatch is avoidable by providing the portion of the amorphous
compound semiconductor layer in between, and occurrence of faults such as dislocation
is avoidable. It is to be noted that the connection layer may be configured, for example,
of the above-described metal atoms, and thereby, contact resistance is sufficiently
reduced, specifically, the contact resistance is allowed to satisfy p-c≤1×10
-3 Ω·cm
2. More specifically, for example, when the connection layer is configured of titanium
(Ti), for example, contact resistance ρ
c with respect to a p
+-GaAs layer and an n
+-InP layer, or with respect to a p
+-GaAs layer and an n
+-InGaAsP layer is allowed to be 1×10
-3 Ω·cm
2 or smaller.
[0036] In the multi-junction solar cell of the present disclosure and the like provided
with the connection layer including the above-described preferable forms and configurations,
the thickness of the amorphous connection layer configured of the above-described
metal or alloy may be desirably 5 nm or smaller, and may be preferably 2 nm or smaller.
FIG. 14 shows a result of measuring a relationship between the thickness of the Ti
layer and transmission characteristics of light having wavelengths within a range
from 450 nm to 800 nm both inclusive, for example, and it can be seen therefrom that
light transmittance of about 80% is secured when the thickness is 5 nm or smaller.
Further, by allowing the thickness to be preferably 2 nm or smaller, light transmittance
of 95% or higher is secured. On the other hand, the thickness of the amorphous connection
layer configured of the above-described materials that is transparent and has electric
conductivity, or the thickness of the amorphous connection layer configured of an
amorphous compound semiconductor may be desirably 1×10
-7 m or smaller.
[0037] When the connection layer is configured of the above-described metal atom, in two
sub-cells that face each other with the connection layer in between (one sub-cell
is referred to as "sub-cell A" for the sake of convenience, and the other sub-cell
is referred to as "sub-cell B" for the sake of convenience), it may be preferable
that a first connection layer is provided in the sub-cell A, a second connection layer
is provided in the sub-cell B, the first connection layer is joined to the second
connection layer to be integrated, and thereby the sub-cell A is joined to the sub-cell
B. In this case, the metal atom configuring the sub-cell A and the metal atom configuring
the sub-cell B may be the same, or may be different. A thickness of the first connection
layer and a thickness of the second connection layer may be the same, or may be different.
It is to be noted that, for example, also when metal or alloy having a work function
that is larger than the Fermi level of the n-type semiconductor or is smaller than
the Fermi level of the p-type semiconductor is used, if the connection layer is configured,
for example, of the first connection layer and the second connection layer having
the same thickness, a width of each depletion layer is 1/2 of that of the pn junction.
Accordingly, probability of causing a tunneling effect is increased. Therefore, it
is an advantageous structure in terms of reduction in contact resistance. It may be
desirable to perform plasma treatment on a junction face of the first connection layer
and a junction face of the second connection layer before joining the first connection
layer to the second connection layer, and thereby, to activate the junction face of
the first connection layer and the junction face of the second connection layer. In
other words, a dangling bond may be desirably formed in the junction interface. Also,
by performing the plasma treatment, the first connection layer and the second connection
layer are allowed to be amorphous. The junction of the first connection layer and
the second connection layer is allowed to be performed at an ambient pressure of 5×10
-4 Pa or lower, with a junction load of 2×10
4 N or less, and at a temperature of 150°C or lower. It may be desirable to perform
the plasma treatment and the junction of the first connection layer and the second
connection layer without breaking a vacuum in terms of prevention of oxidation of
the junction face of the first connection layer and the junction face of the second
connection layer.
[0038] Moreover, in the multi-junction solar cell of the present disclosure including the
above-described preferable forms and configurations, it may be preferable to adopt
a form in which conductivity types of the compound semiconductor layers that face
each other in the sub-cells adjacent to each other are different. In particular, where
the sub-cells adjacent to each other are "sub-cell a" and "sub-cell b", a compound
semiconductor layer, in the sub-cell a, that faces the sub-cell b is "compound semiconductor
layer a", and a compound semiconductor layer, in the sub-cell b, that faces the sub-cell
a is "compound semiconductor layer b", it may be preferable to adopt a form in which
a conductivity type of the compound semiconductor layer a is different from a conductivity
type of the compound semiconductor layer b. Alternatively, it may be preferable to
adopt a form in which conductivity types of the compound semiconductor layers that
face each other with the connection layer in between are different. The same is applicable
also to the photoelectric conversion device of the present disclosure including the
above-described various preferable forms and configurations.
[0039] Moreover, in the multi-junction solar cell of the present disclosure including the
above-described preferable forms and configurations, it may be desirable to adopt
a form in which a thickness of a compound semiconductor layer having the conductivity
type of the p-type out of the compound semiconductor layers configuring the sub-cells,
more specifically, a thickness of the p
+-GaAs layer is 100 nm or smaller. The same is applicable also to the photoelectric
conversion device of the present disclosure including the above-described various
preferable forms and configurations.
[0040] Moreover, in the multi-junction solar cell of the present disclosure including the
above-described preferable forms and configurations, as the compound semiconductor
that configures a sub-cell other than the predetermined sub-cell, or as the compound
semiconductor layer that configures other photoelectric conversion device in the laminated
photoelectric conversion devices, an InGaAs layer, an InGaAsP layer, a GaAs layer,
an InGaP layer, an AlInGaP layer, a GaAsN layer, an InGaAsN layer, an InP layer, an
InAlAs layer, an InAlAsSb layer, an InGaAlAs layer, and an AlGaAs layer can be mentioned.
[0041] More specifically, when the multi-junction solar cell of the present disclosure including
the above-described preferable forms and configurations is configured of two sub-cells,
the respective sub-cells may be configured, for example, of the followings.
(InGaAsP layer, InGaAs layer)
(InGaAs layer, InGaAs layer)
(InP layer, InGaAs layer)
[0042] It is to be noted that light enters from the sub-cell having a layer configuration
described on the most-left side in (). Further, when three sub-cells are used, the
respective sub-cells may be configured, for example, of the followings.
(GaAs layer, InGaAsP layer, InGaAs layer)
(InGaAs layer, InGaAsP layer, InGaAs layer)
(InGaP layer, InGaAs layer, InGaAs layer)
[0043] Further, when four sub-cells are used, the respective sub-cells may be configured,
for example, of the followings.
(GaInP layer, GaAs layer, InGaAsP layer, InGaAs layer)
(GaInP layer, InGaAs layer, InGaAsP layer, InGaAs layer)
(GaInP layer, InGaAs layer, InGaAsN layer, InGaAs layer)
[0044] Further, when five sub-cells are used, the respective sub-cells may be configured,
for example, of the followings.
(GaInP layer, GaAs layer, InGaAs layer, InGaAsP layer, InGaAs layer)
(GaInP layer, GaAs layer, InGaAsN layer, InGaAsP layer, InGaAs layer)
(GaInP layer, GaAs layer, InGaAs layer, InGaAs layer, InGaAs layer)
[0045] Further, when six sub-cells are used, the respective sub-cells may be configured,
for example, of the following.
(AIGaInP, GaInP, AlGaInAs, GaAs, InGaAs, InGaAs)
[0046] It is to be noted that, when a plurality of sub-cells are described to be configured
of the same compound semiconductor in one multi-junction solar cell, composition ratios
thereof are different.
[0047] The multi-junction solar cell, the photoelectric conversion device, or the compound-semiconductor-layer
lamination structure of the present disclosure including the above-described preferable
forms and configurations is provided on a substrate. A substrate for film formation
that is used at the time of manufacturing the multi-junction solar cell, the photoelectric
conversion device, or the compound-semiconductor-layer lamination structure of the
present disclosure and a support substrate may be the same substrate, or may be different
substrates. It is to be noted that a substrate (corresponding to the base) in a case
where the substrate for film formation is the same as the support substrate is expressed
as "substrate for film-formation/support" for the sake of convenience. Alternatively,
when the substrate for film formation is different from the support substrate, the
respective substrates are expressed as "substrate for film formation" and "support
substrate". In this case, after the compound semiconductor layers and the like are
formed on the substrate for film formation (corresponding to the base), the substrate
for film formation may be removed from the compound semiconductor layers and the like,
and the compound semiconductor layers and the like may be fixed to the support substrate,
or may be bonded to the support substrate. As a method of removing the substrate for
film formation from the compound semiconductor layers and the like, a laser ablation
method, a heating method, an etching method, etc. can be mentioned. Further, as a
method of fixing or bonding the compound semiconductor layers and the like to the
support substrate, a metal junction method, a semiconductor junction method, and a
metal-semiconductor junction method may be mentioned other than a method using a bonding
agent.
[0048] As described above, the substrate for film-formation/support may be preferably formed
of InP. In other words, the substrate for film-formation/support may be preferably
configured of an InP substrate. Further, as described above, the substrate for film
formation used for manufacturing the predetermined sub-cell in the multi-junction
solar cell of the present disclosure and the substrate for film formation used for
manufacturing the photoelectric conversion device of the present disclosure may be
preferably formed of InP. In other words, the substrate for film formation used for
manufacturing the predetermined sub-cell in the multi-junction solar cell of the present
disclosure and the substrate for film formation used for manufacturing the photoelectric
conversion device of the present disclosure may be preferably configured of InP substrates.
[0049] On the other hand, as the substrate for film formation used for manufacturing the
sub-cell other than the predetermined sub-cell in the multi-junction solar cell of
the present disclosure, or the substrate for film formation used for manufacturing
the other photoelectric conversion device in the photoelectric conversion device that
has a lamination structure in which the photoelectric conversion device of the present
disclosure and other photoelectric conversion device are laminated, a substrate made
of a III-V group semiconductor or a II-VI group semiconductor can be mentioned. Specifically,
as the substrate made of the III-V group semiconductor, GaAs, InP, GaN, AlN, etc.
can be mentioned, and as the substrate made of the II-VI group semiconductor, CdS,
CdTe, ZnSe, ZnS, etc. can be mentioned. Further, a substrate made of a I-III-VI group
semiconductor called a chalcopyrite type made of Cu, In, Ga, Al, Se, S, etc. may be
used, and specifically, Cu(In,Ga)Se
2 abbreviated as CIGS, Cu(In,Ga)(Se,S)
2 abbreviated as CIGSS, CuInS
2 abbreviated as CIS, etc. can be mentioned.
[0050] Moreover, as the support substrate, other than the above-described various substrates,
a glass substrate, a quartz substrate, a transparent inorganic substrate such as a
sapphire substrate, and a transparent plastic substrate or film made of a material
such as: polyester resin such as polyethylene terephthalate (PET) and polyethylene
naphthalate (PEN); polycarbonate (PC) resin; polyether sulfone (PES) resin; polyolefin
resin such as polystyrene, polyethylene, and polypropylene; polyphenylene sulfide
resin; polyvinylidene difluoride resin; tetraacetyl cellulose resin; phenoxy bromide
resin; aramid resin; polyimide resin; polystyrene resin; polyarylate resin; polysulfone
resin; acrylic resin; epoxy resin; fluororesin; silicone resin; diacetate resin; triacetate
resin; polyvinyl chloride resin; and cyclic polyolefin resin can be mentioned. As
the glass substrate, for example, a soda glass substrate, a heat-resistance glass
substrate, and a quartz glass substrate can be mentioned.
[0051] A second electrode is formed on a sub-cell in an uppermost layer in the sub-cells
configuring the multi-junction solar cell of the present disclosure. The second electrode
may preferably have a thickness, for example, of about 10 nm to 100 nm both inclusive,
and may be preferably configured of a material having favorable light transmission
characteristics and having a small work function. As such a material, for example,
indium-tine oxide (including ITO, Indium Tin Oxide, Sn-doped In
2O
3, crystalline ITO, and amorphous ITO), indium-zinc oxide (IZO, Indium Zinc Oxide),
IFO (F-doped In
2O
3), tin oxide (SnO
2), ATO (Sb-doped SnO
2), FTO (F-doped SnO
2), zinc oxide (including ZnO, Al-doped ZnO, and B-doped ZnO), InSnZnO, spinel-type
oxides, oxides having a YbFe
2O
4 structure, etc. can be mentioned. Further, alkaline-earth metal such as calcium (Ca)
and barium (Ba), alkali metal such as lithium (Li) and cesium (Cs), indium (In), magnesium
(Mg), silver (Ag), gold (Au), nickel (Ni), gold-germanium (Au-Ge), etc. can also be
mentioned. Further, alkali-metal oxide such as Li
2O, Cs
2Co
3, Cs
2SO
4, MgF, LiF, and CaF
2, alkali-metal fluoride, alkaline-earth-metal oxide, and alkaline-earth fluoride can
also be mentioned. The second electrode may have a single-layer configuration, or
may have a configuration made by laminating a plurality of layers. The second electrode
is allowed to be formed by a physical vapor deposition method (a PVD method) such
as a vacuum evaporation method and a sputtering method, a chemical vapor deposition
method (a CVD method), etc. The same is applicable also to the photoelectric conversion
device of the present disclosure.
[0052] The first electrode is formed on the sub-cells, the compound semiconductor layers,
etc. Alternatively, depending on materials configuring the substrate for film-formation/support
or the support substrate, the substrate for film-formation/support or the support
substrate itself may be used as the first electrode. As a material configuring the
first electrode, molybdenum (Mo), tungsten (W), tantalum (Ta), vanadium (V), palladium
(Pd), zinc (Zn), nickel (Ni), titanium (Ti), platinum (Pt), and gold-zinc (Au-Zn)
can be mentioned as examples.
[0053] An anti-reflection film may be preferably formed on a portion, on the sub-cell in
the uppermost layer (the sub-cell on the light incident side) in the sub-cells configuring
the multi-junction solar cell of the present disclosure, on which the second electrode
is not formed. The anti-reflection film is provided in order to suppress reflection
in the sub-cell in the uppermost layer, and to take in solar light efficiently in
the multi-junction solar cell of the present disclosure. As a material configuring
the anti-reflection film, a material having a refractive index that is smaller than
that of the compound semiconductor configuring the sub-cell in the uppermost layer
may be preferably used. Specifically, for example, a layer made of TiO
2, Al
2O
3, ZnS, MgF
2, Ta
2O
5, SiO
2, or Si
3N
4, or a lamination structure of these layers can be mentioned. As a film thickness
of the anti-reflection film, for example, from 10 nm to 200 nm both inclusive can
be mentioned. The same is applicable also to the photoelectric conversion device of
the present disclosure.
EXAMPLE 1
[0054] Example 1 relates to the multi-junction solar cell, the photoelectric conversion
device, and the compound-semiconductor-layer lamination structure of the present disclosure.
(A) of FIG. 1 shows a conceptual diagram of a multi-junction solar cell in Example
1. (B) of FIG. 1 shows a conceptual diagram of a predetermined sub-cell located in
a lowermost layer, a photoelectric conversion device, and a compound-semiconductor-layer
lamination structure.
[0055] The multi-junction solar cell in Example 1 shown in the conceptual diagram in (A)
of FIG. 1 is a multi-junction solar cell in which a plurality of (four, in Example
1) sub-cells (a first sub-cell 11, a second sub-cell 12, a third sub-cell 13, and
a fourth sub-cell 14) are laminated, light enters from the fourth sub-cell 14 located
in an uppermost layer to the first sub-cell 11 located in the lowermost layer, and
electric power is generated in each of the sub-cells 11, 12, 13, and 14. Each of the
sub-cells 11, 12, 13, and 14 is configured of the first compound semiconductor layer
having the first conductivity type and the second compound semiconductor layer having
the second conductivity type that are laminated. It is to be noted that, in the description
below, the first conductivity type is set as a p-type, and the second conductivity
type is set as an n-type. Further, the base, the substrate for film-formation/support,
and the substrate for film formation are each configured of a p-type InP substrate.
However, the present disclosure is not limited thereto.
[0056] A lamination order in the plurality of sub-cells is set to be a lamination order
by which a band gap of the compound semiconductor configuring the sub-cell becomes
larger as the compound semiconductor is closer to the light incident side, that is,
a lamination order by which the band gaps of the compound semiconductors are larger
in order from a substrate for film-formation/support 31 side to a second electrode
side. Specifically, on the substrate for film-formation/support 31, the first sub-cell
11, the second sub-cell 12, the third sub-cell 13, and the fourth sub-cell 14 are
formed in this order, and for example, solar light may enter from the fourth sub-cell
14.
[0057] Further, the photoelectric conversion device in Example 1 is a photoelectric conversion
device configured of a first compound semiconductor layer 11A having the first conductivity
type and a second compound semiconductor layer 11C having the second conductivity
type that are laminated.
[0058] It is to be noted that, hereinbelow, mostly, detailed description will be provided
on at least one predetermined sub-cell (the sub-cell 11 located in the lowermost layer,
in Example 1) out of the plurality of sub-cells. Detailed description will be provided
on the remaining sub-cells 12, 13, and 14 from Example 3 and thereafter.
[0059] Further, in at least one predetermined sub-cell (the sub-cell 11 located in the lowermost
layer, in Example 1) out of the plurality of sub-cells, or in the photoelectric conversion
device in Example 1, the first compound semiconductor layer 11A is configured of at
least one first-compound-semiconductor-layer lamination unit 11A
U in which a 1-A compound semiconductor layer 11A
A of a p-type and a 1-B compound semiconductor layer 11A
B of a p-type are laminated. Further, the second compound semiconductor layer 11C is
configured of at least one second-compound-semiconductor-layer lamination unit 11C
U in which a 2-A compound semiconductor layer 11C
A of an n-type (more specifically, an n
+-type, in Example 1) and a 2-B compound semiconductor layer 11C
B of an n-type (more specifically, an n
+-type, in Example 1) are laminated. It is to be noted that, in (B) of FIG. 1, three
first-compound-semiconductor-layer lamination units 11A
U and three second-compound-semiconductor-layer lamination units 11C
U are illustrated; however, this is not limitative. Moreover, a compound semiconductor
composition configuring the 1-A compound semiconductor layer 11A
A and a compound semiconductor composition configuring the 2-A compound semiconductor
layer 11C
A are the same compound semiconductor composition A. Further, a compound semiconductor
composition configuring the 1-B compound semiconductor layer 11A
B and a compound semiconductor composition configuring the 2-B compound semiconductor
layer 11C
B are the same compound semiconductor composition B. In other words, a group of atoms
configuring the compound semiconductor composition A and a group of atoms configuring
the compound semiconductor composition B are the same, and are specifically, InGaAs.
However, as will be described later, atomic percentage of the group of atoms configuring
the compound semiconductor composition A is different from atomic percentage of the
group of atoms configuring the compound semiconductor composition B.
[0060] Band gaps of the second sub-cell and the first sub-cell were variously varied, simulation
is performed, and conversion efficiency was determined when the AM 1.5 solar light
was in a non-condensed state (1 sun), and the followings were set.
a band gap of the fourth sub-cell: 1.910 eV
a band gap of the third sub-cell: 1.420 eV
[0061] Result thereof is shown in FIG. 12, which shows that a region in which the conversion
efficiency is high is present in the following range.
the band gap of the first sub-cell: from 0.46 eV to 0.56 eV both inclusive
the band gap of the second sub-cell: from 1.01 eV to 1.04 eV both inclusive
[0062] Moreover, as a result of the simulation, when
the band gap of the first sub-cell: 0.536 eV
the band gap of the second sub-cell: 1.020 eV
were established,
FF=87.6%
VOC=3.358 volts
Vmp=3.020 volts
Jmp=13.807 milliamperes
were obtained, and 41.64% (non-condensed light) was obtained as maximum conversion
efficiency. It is to be noted that, in FIG. 12, the conversion efficiency is increased
from a region indicated by an arrow "A" to a region indicated by an arrow "C", the
conversion efficiency is increased from a region indicated by an arrow "B" to the
region indicated by the arrow "C", and the region indicated by the arrow "C" is a
region with highest conversion efficiency.
[0063] Further, the compound semiconductor composition A is determined based on a value
of a band gap of the predetermined sub-cell 11, or based on a value of a band gap
of the photoelectric conversion device. Here, the value of the band gap of the predetermined
sub-cell 11 or the value of the band gap of the photoelectric conversion device is
from 0.45 eV to 0.75 eV both inclusive. However, in Example 1, more specifically,
the value of the band gap of the predetermined sub-cell 11 or the value of the band
gap of the photoelectric conversion device was set to 0.65 eV. A compound semiconductor
layer that has such a desirable value of band gap favorably absorbs light with a wavelength
of 1907.7 nm or less. Further, the compound semiconductor composition A that is the
composition of the 1-A compound semiconductor layer that has such a value of band
gap is In
0.63Ga
0.37As. It is to be noted that a critical film thickness of In
0.63Ga
0.37As is 72 nm. Further, when the film thickness is 12 nm or smaller, quantum effect
is remarkably caused.
[0064] Where LC
A is the lattice constant of the compound semiconductor composition A, LC
A is the lattice constant of the compound semiconductor composition B, and LC
0 is the base lattice constant, values of LC
A, LC
B, LC
0, and (LC
A-LC
0)/LC
0 are as shown in Table 1 below.
[0065] In Example 1, the compound semiconductor composition B is determined based on a difference
between the base lattice constant LC
0 of the base (specifically, the p-type InP substrate) used at a time of forming the
first compound semiconductor layer 11A and the second compound semiconductor layer
11C, and the lattice constant LC
A of the compound semiconductor composition A. Specifically, as shown in Table 1, since

is established, the compound semiconductor composition B is determined so that a value
of (LC
B-LC
0)/LC
0 which cancels this value of (LC
A-LC
0)/LC
0 is obtained. More specifically, the composition of the compound semiconductor composition
B was set as In
0.45Ga
0.55As. It is to be noted that a value of (LC
B-LC
0)/LC
0 is as shown in Table 1 below. Further, a value of a band gap of the compound semiconductor
composition B is larger than a value of a band gap of the compound semiconductor composition
A.
[0066] As described above,

and

are satisfied, and further,

and

are established.
[0067] Moreover, a thickness t
B of the 1-B compound semiconductor layer is determined based on a difference between
the base lattice constant LC
0 and a lattice constant LC
B of the compound semiconductor composition B, and on a thickness t
A of the 1-A compound semiconductor layer. Further, a thickness t
B of the 2-B compound semiconductor layer is determined based on the difference between
the base lattice constant LC
0 and the lattice constant LC
B of the compound semiconductor composition B, and on a thickness t
A of the 2-A compound semiconductor layer. The thickness t
A of the 1-A compound semiconductor layer and the 2-A compound semiconductor layer
and the thickness t
B of the 1-B compound semiconductor layer and the 2-B compound semiconductor layer
satisfy the following expression.

[0068] Here, the thickness t
A of the 1-A compound semiconductor layer and the 2-A compound semiconductor layer
is smaller than a critical film thickness of the compound semiconductor composition
A, and is a thickness that causes no quantum effect. Further, the thickness t
B of the 1-B compound semiconductor layer and the 2-B compound semiconductor layer
is smaller than a critical film thickness of the compound semiconductor composition
B, and is a thickness that causes no quantum effect. It is to be noted that, in the
compound semiconductor composition B, when the film thickness is 15 nm or smaller,
quantum effect is remarkably caused.
[0069] The above-described various values are summarized in Table 1 and Table 2 below.
[Table 1]
LCA=5.908 Å |
LCB=5.836 Å |
LC0=5.868 Å |
(LCA-LC0)/LC0=6.8 × 10-3 |
(LCB-LC0)/LC0=-5.5 × 10-3 |
[Table 2]
Compound semiconductor composition A: |
Composition: |
In0.63Ga0.37As (x=0.63) |
Critical film thickness |
72 nm |
Band gap: |
0.65 eV |
Compound semiconductor composition B: |
Composition: |
In0.45Ga0.55As (y=0.45) |
Critical film thickness |
94.5 nm |
Band gap: |
0.83 eV |
Thickness tA of the 1-A compound semiconductor layer and the 2-A compound semiconductor layer: |
50 nm |
Thickness tB of the 1-B compound semiconductor layer and the 2-B compound semiconductor layer: |
40 nm |
[0070] Here, a total thickness of the first compound semiconductor layers 11 A having the
conductivity type of a p-type was set as 3.0 µm. Therefore, the number of the first-compound-semiconductor-layer
lamination units 11A
U in which the 1-A compound semiconductor layer 11A
A and the 1-B compound semiconductor layer 11A
B are laminated is 33. It is to be noted that, the first compound semiconductor layers
11A having the conductivity type of a p-type has a lamination structure that has a
conductivity type of a p
+-type on an InP substrate side, and has a conductivity type of a p-type on a second
compound semiconductor layer 11C side. A total thickness of the first compound semiconductor
layers 11A
1 having the conductivity type of a p
+-type was set as 0.1 µm, and a total thickness of the first compound semiconductor
layers 11A
2 having the conductivity type of a p-type was set as 2.9 µm. Further, a total thickness
of the second compound semiconductor layers 11C having a conductivity type of an n
+-type was set as 0.2 µm. Therefore, the number of the second-compound-semiconductor-layer
lamination units 11C
U in which the 2-A compound semiconductor layer 11C
A and the 2-B compound semiconductor layer 11C
B are laminated is from 2 to 3 both inclusive.
[0071] The thickness of the first compound semiconductor layer 11A is a product of (the
number of the first-compound-semiconductor-layer lamination units 11A
U) and (t
A+t
B). It may be desirable to allow a value of (t
A+t
B) to be as large as possible, and to allow a value of (the number of the first-compound-semiconductor-layer
lamination units 11A
U) to be as small as possible, for example, in terms of reduction in the number of
energy gaps formed in the first compound semiconductor layer 11A. Various tests may
be performed on the multi-junction solar cell or the photoelectric conversion device,
and the values of (the number of the first-compound-semiconductor-layer lamination
units 11A
U) and (t
A+t
B) may be determined based thereon. The same is applicable also to values of (the number
of the second-compound-semiconductor-layer lamination units 11C
U) and (t
A+t
B). Also, the same is applicable to the Examples below as well.
[0072] The compound-semiconductor-layer lamination structure in Example 1 is a compound-semiconductor-layer
lamination structure that includes at least one compound-semiconductor-layer lamination
unit in which the A compound semiconductor layer 11A
A and a B compound semiconductor layer 11A
B are laminated, wherein
the compound semiconductor composition B configuring the B compound semiconductor
layer 11A
B is determined based on a difference between the base lattice constant LC
0 of the base (specifically, InP) used at a time of forming the A compound semiconductor
layer 11A
A and the B compound semiconductor layer 11A
B and the lattice constant LC
A of the compound semiconductor composition A configuring the A compound semiconductor
layer 11A
A,
a thickness of the B compound semiconductor layer 11A
B is determined based on a difference between the base lattice constant LC
0 and the lattice constant LC
B of the compound semiconductor composition B,
the thickness of the A compound semiconductor layer 11A
A is smaller than the critical film thickness of the compound semiconductor composition
A, and is a thickness that causes no quantum effect, and
the thickness of the B compound semiconductor layer 11A
B is smaller than the critical film thickness of the compound semiconductor composition
B, and is a thickness that causes no quantum effect.
[0073] As described above, the sub-cell 11 that configures the lowermost layer in the multi-junction
solar cell in Example 1 favorably absorbs light with a wavelength of 1907.7 nm or
less. Further, the sub-cell 11 as a whole is lattice-matched with the InP substrate
which is the base.
[0074] In the multi-junction solar cell in Example 1, when a band gap of the predetermined
sub-cell 11, in other words, for example, a wavelength of light which the predetermined
sub-cell 11 is allowed to absorb most efficiently is set, or when a band gap of the
photoelectric conversion device, in other words, a wavelength of light which the photoelectric
conversion device is allowed to absorb most efficiently or a desirable light emission
wavelength is set, the compound semiconductor composition A that achieves this is
determined. However, usually, the lattice constant LC
A of the determined compound semiconductor composition A and the base lattice constant
LC
0 have a difference, in other words, often, lattice-unmatched type is established.
Therefore, in order to eliminate this difference, in other words, in order to cancel
this difference (in order to establish a lattice-matched type), the compound semiconductor
composition B is determined. Moreover, based on the difference between the base lattice
constant LC
0 and the lattice constant LC
B of the compound semiconductor composition B, and on the thickness t
A of the 1-A compound semiconductor layer and the 2-A compound semiconductor layer,
not only the thickness t
B of the 1-B compound semiconductor layer and the 2-B compound semiconductor layer
is determined, but also, upper-limit values and lower-limit values of layer thicknesses
t
A and t
B of the 1-A and 2-A compound semiconductor layers and the 1-B and 2-B compound semiconductor
layers are defined.
[0075] In such a manner, in each of the first compound semiconductor layer and the second
compound semiconductor layer, the compound semiconductor composition is optimized
(in other words, the band gap and the lattice constant are optimized), and the layer
thickness is optimized. As a result, even when lattice mismatch is established between
the 1-A compound semiconductor layer and the base, this lattice mismatch is cancelled
by the 1-B compound semiconductor layer, and the first compound semiconductor layer
as a whole becomes of a lattice-matched type. In other words, the 1-A compound semiconductor
layer and the 1-B compound semiconductor layer configure a distortion compensation
lamination structure. The same is applicable also to the second compound semiconductor
layer as a whole. Moreover, the first compound semiconductor layer and the second
compound semiconductor layer as a whole are allowed to achieve efficient absorption
with a desirable wavelength of light or light emission with a desirable wavelength.
Further, this may, for example, further improve efficiency in absorption of solar
light having a wide wavelength range in the multi-junction solar cell.
[0076] Moreover, in the compound-semiconductor-layer lamination structure in Example 1 the
compound semiconductor composition B is determined based on the difference between
the base lattice constant LC
0 and the lattice constant LC
A of the compound semiconductor composition A, and the thickness of the B compound
semiconductor layer 11A
B is determined based on the difference between the base lattice constant LC
0 and the lattice constant LC
B of the compound semiconductor composition B and on the thickness t
A of the A compound semiconductor layer 11A
A. Therefore, even when lattice mismatch is established between the A compound semiconductor
layer 11A
A and the base, this lattice mismatch is cancelled by the B compound semiconductor
layer 11A
B, and the compound-semiconductor-layer lamination structure as a whole becomes of
a lattice-matched type. Therefore, restriction such as that it is necessary to determine
the compound semiconductor composition A being limited by the base lattice constant
in order to obtain the lattice-matched type is eased. Therefore, it is possible to
expand the range of options for the compound semiconductor composition of the compound
semiconductor layer configuring the compound-semiconductor-layer lamination structure,
and to improve freedom in options.
EXAMPLE 2
[0077] Example 2 is a modification of Example 1. In Example 2, the value of the band gap
of the predetermined sub-cell 11, or the value of the band gap of the photoelectric
conversion device was set as 0.55 eV. A compound semiconductor layer that has such
a value of band gap favorably absorbs light with a wavelength of 2254.5 nm or less.
Further, the compound semiconductor composition A that is a composition of the 1-A
compound semiconductor layer that has such a value of band gap is In
0.74Ga
0.26As. It is to be noted that a critical film thickness of In
0.74Ga
0.26As is 55.5 nm. Further, when the film thickness is 12 nm or smaller, quantum effect
is remarkably caused.
[0078] Moreover, where LC
A is the lattice constant of the compound semiconductor composition A, LC
A is the lattice constant of the compound semiconductor composition B, and LC
0 is the base lattice constant, values of LC
A, LC
B, LC
0, and (LC
A-LC
0)/LC
0 are as shown in Table 3 below. Also, the composition of the compound semiconductor
composition A, the composition of the compound semiconductor composition B, the thickness
t
A of the 1-A compound semiconductor layer and the 2-A compound semiconductor layer,
and the thickness t
B of the 1-B compound semiconductor layer and the 2-B compound semiconductor layer
are shown in Table 4. It is to be noted that, in the compound semiconductor composition
B, when the film thickness is 15 nm or smaller, quantum effect is remarkably caused.
[Table 3]
LCA=5.953 Å |
LCB=5.783 Å |
LC0=5.868 Å |
(LCA-LC0)/LC0=1.4×10-2 |
(LCB-LC0)/LC0=-1.4×10-2 |
[Table 4]
Compound semiconductor composition A: |
Composition: |
In0.74Ga0.26As (x=0.74) |
Critical film thickness |
28.5 nm |
Band gap: |
0.55 eV |
Compound semiconductor composition B: |
Composition: |
In0.32Ga0.68As (y=0.32) |
Critical film thickness |
27.5 nm |
Band gap: |
0.98 eV |
Thickness tA of the 1-A compound semiconductor layer and the 2-A compound semiconductor layer: |
24 nm |
Thickness tB of the 1-B compound semiconductor layer and the 2-B compound semiconductor layer: |
24 nm |
[0079] Here, the total thickness of the first compound semiconductor layers 11A having the
conductivity type of a p-type was set as 3.0 µm. Therefore, the number of the first-compound-semiconductor-layer
lamination units 11A
U in which the 1-A compound semiconductor layer 11A
A and the 1-B compound semiconductor layer 11A
B are laminated is 63. It is to be noted that, the first compound semiconductor layers
11A having the conductivity type of a p-type has a lamination structure that has a
conductivity type of a p
+-type on an InP substrate side, and has a conductivity type of a p-type on a second
compound semiconductor layer 11C side. The total thickness of the first compound semiconductor
layers 11A
1 having the conductivity type of a p
+-type was set as 0.1 µm, and the total thickness of the first compound semiconductor
layers 11A
2 having the conductivity type of a p-type was set as 2.9 µm. Further, the total thickness
of the second compound semiconductor layers 11C having a conductivity type of an n
+-type was set as 0.2 µm. Therefore, the number of the second-compound-semiconductor-layer
lamination units 11C
U in which the 2-A compound semiconductor layer 11C
A and the 2-B compound semiconductor layer 11C
B are laminated is from 4 to 5 both inclusive.
EXAMPLE 3
[0080] Description on the predetermined sub-cell 11 has been given in Example 1 and Example
2. Description will be given on other sub-cells and the connection layer in Examples
below.
[0081] Each of the sub-cells 11, 12, 13, and 14 is formed of a plurality of a plurality
of compound semiconductor layers that are laminated. A configuration of each of the
sub-cells 11, 12, 13, and 14 is shown in Table 5 below. It is to be noted that, in
Table 5, concerning the compound semiconductor layers configuring each sub-cell, a
compound semiconductor layer closer to the support substrate is described on a lower
side, and a compound semiconductor layer farther from the support substrate is described
on an upper side. Moreover, an amorphous connection layer 20 (connection layers 20A
and 20B) made of an electrically-conductive material is provided in at least one place
between the adjacent sub-cells, that is, between the second sub-cell 12 and the third
sub-cell 13 that are of the lattice-mismatched type in Example 3. Here, the connection
layer 20 is made of titanium (Ti) having a thickness of 1.0 nm. It is to be noted
that the connection layer 20 has the two-dimensional layer structure and does not
have the three-dimensional island structure.
[Table 5]
Fourth sub-cell 14: Band gap 1.90 eV, Lattice constant 5.653 Å |
Compound semiconductor layer 14C: n+-In0.48Ga0.52P |
Compound semiconductor layer 14B: p-In0.48Ga0.52 |
Compound semiconductor layer 14A: p+-In0.48Ga0.52 |
Third sub-cell 13: Band gap 1.42 eV, Lattice constant 5.653 Å |
Compound semiconductor layer 13C: n+-GaAs |
Compound semiconductor layer 13B: p-GaAs |
Compound semiconductor layer 13A: p+-GaAs |
Second sub-cell 12: Band gap 1.02 eV, Lattice constant 5.868 Å |
Compound semiconductor layer 12C: n+-In0.79Ga0.21As0.43PO057 |
Compound semiconductor layer 12B: p-In0.79Ga0.21As0.43P0.57 |
Compound semiconductor layer 12A: p+-In0.79Ga0.21As0.43P0.57 |
First sub-cell 11: See Example 1 or Example 2
[0082] Moreover, in the multi-function solar cell in Example 3, for example, a second electrode
19 configured of a lamination structure of AuGe/Ni/Au having thicknesses of 150 nm/50
nm/500 nm is formed on the fourth sub-cell 14, and an anti-reflection film 18 configured
of a TiO
2 film and an Al
2O
3 film is formed on a portion, on the fourth sub-cell 14, on which the second electrode
19 is not formed. It is to be noted that, in the drawings, the second electrode 19
and the anti-reflection film 18 are each illustrated as one layer. The substrate for
film-formation/support 31 is configured of a p-type InP substrate. Further, a first
tunnel junction layer 15 configured of p
+-InGaAs (upper layer)/n
+-InGaAs (lower layer) is provided between the first sub-cell 11 and the second sub-cell
12 that are of the lattice-matched type, and a second tunnel junction layer 16 configured
of p
+-InGaP (upper layer)/n
+-InGaP (lower layer) is provided between the third sub-cell 13 and the fourth sub-cell
14 that are of the lattice-matched type. Further, a window layer 17 configured of
n
+-AlInP is formed between the fourth sub-cell 14, and the second electrode 19 and the
anti-reflection film 18. It is to be noted that the window layer 17 is provided in
order to prevent recombination of carriers in the uppermost surface, but may not be
necessarily provided. The first electrode is connected to the first sub-cell 11; however,
illustration of the first electrode is omitted.
[0083] Description will be given below of a method of manufacturing the multi-junction solar
cell, the photoelectric conversion device, and the compound-semiconductor-layer lamination
structure in Example 3 referring to (A) of FIG. 1, (A) to (B) of FIG. 2, and (A) to
(B) of FIG. 3 that are conceptual diagrams of the compound semiconductor layers and
the like.
[Step-300]
[0084] On the substrate for film-formation/support 31 configured of the p-type InP substrate,
the first sub-cell 11 (the compound semiconductor layers 11A
1, 11A
2, and 11C), the first tunnel junction layer 15, and the second sub-cell 12 (the compound
semiconductor layers 12A to 12C) that are of the lattice-matched type are epitaxially
grown in a sequential manner based on an MOCVD method. On the other hand, on a substrate
for film formation 44 configured of an n-type GaAs substrate, a sacrificial layer
for peeling-off 45 configured of AlAs is formed, and then, the window layer 17 configured
of n
+-AlInP is formed, based on the MOCVD method. Subsequently, on this window layer 17,
the fourth sub-cell layer 14 (the compound semiconductor layers 14C to 14A), the second
tunnel junction layer 16, and the third sub-cell 13 (the compound semiconductor layers
13C to 13A) that are of the lattice-matched type are epitaxially grown in a sequential
manner. Thus, a structure shown in the conceptual diagram in (A) of FIG. 2 is obtained.
[0085] Next, the compound semiconductor layer 12C configured of n
+-In
0.79Ga
0.21As
0.43P
0.57 configuring the second compound semiconductor layer 12 is joined to the compound
semiconductor layer 13A configured of p
+-GaAs configuring the third compound semiconductor layer 13 with the connection layer
20 in between, and thereby, ohmic contact is obtained.
[Step- 310]
[0086] Specifically, first, the first connection layer 20A is formed on the compound semiconductor
layer 12C configuring the second compound semiconductor layer 12, and the second connection
layer 20B is formed on the compound semiconductor layer 13A configuring the third
compound semiconductor layer 13 (see (B) of FIG. 2). More specifically, for example,
the connection layers 20A and 20B each configured of Ti with a film thickness of 0.5
nm may be formed on each of the compound semiconductor layer 12C and the compound
semiconductor layer 13A based on a vacuum evaporation method (under conditions of:
at a vacuum degree of 2×10
-4 Pa; at evaporation speed of 0.1 nm/sec or lower; and at a temperature from 150°C
to 200°C both inclusive). It is to be noted that, in this case, for example, a resistance
heating scheme may be adopted setting a substrate temperature at 80°C and a substrate
rotation speed at 30 rpm. However, the film formation method of the connection layers
20A and 20B is not limited thereto, and, for example, a sputtering method (under conditions
of: at film formation speed of 0.1 nm/sec or lower; and at a temperature from 150°C
to 200°C both inclusive) may be used.
[Step-320]
[0087] Subsequently, after performing plasma treatment on the connection layers 20A and
20B, the second compound semiconductor layer 12 is joined to the third compound semiconductor
layer 13. Specifically, argon (Ar) plasma (for example, at plasma density from 10
9 cm
-3 to 10
11 cm
-3 both inclusive, and at a pressure from 1 Pa to 10
-2 Pa both inclusive) is applied to surfaces of the connection layers 20A and 20B, and
thereby, the surfaces (junction faces) of the connection layers 20A and 20B are activated.
In other words, a dangling bond is formed in the junction interface (the surfaces
of the connection layers 20A and 20B). Also, the connection layers 20A and 20B are
allowed to be amorphous. Further, the connection layers 20A and 20B are joined (bonded)
to each other while maintaining high degree of vacuum, that is, at an ambient pressure
of 5×10
-4 Pa or lower, at junction load of 2×10
4 N or less, and at a temperature of 150°C or lower, specifically, for example, at
an ambient pressure of 1×10
-4 Pa, at junction load of 2×10
4 N, and at a temperature of 25°C. Thus, a structure illustrated in the conceptual
diagram in (A) of FIG. 3 is obtained. In Example 3, metal (specifically, Ti) is used
as the material of the connection layer 20. As described above, at the time of film
formation, the metal thin film may be often formed in an island-like form, and a layer-like
form is rarely obtained. However, film formation in the layer-like form is possible
with the metal atoms in Group (A) and Group (B) shown in FIG. 13.
[Step-330]
[0088] Thereafter, the substrate for film formation 44 is peeled off, and the anti-reflection
film 18 and the second electrode 19 are formed. Specifically, after the substrate
for film formation 44 is peeled off by removing the sacrificial layer for peeling-off
45 is removed by etching (see (B) of FIG. 3), for example, a resist pattern may be
formed on the window layer 17 based on a photolithography technique, and the second
electrode 19 may be formed by a vacuum evaporation method (at vacuum degree of 2×10
-4 Pa, at evaporation speed of 0.1 nm/sec, and at a temperature from 150°C to 200°C
both inclusive). It is to be noted that the substrate for film formation 44 is allowed
to be reused. Next, by removing the resist pattern, the second electrode 19 is formed
based on a lift-off method. Subsequently, a resist pattern is formed based on a photolithography
technique, and, for example, the anti-reflection film 18 configured of a TiO
2 film and an Al
2O
3 film may be formed by a vacuum evaporation method (at vacuum degree of 2×10
-4 Pa, at evaporation speed of 0.1 nm/sec, and at a temperature from 150°C to 200°C
both inclusive). Subsequently, by removing the resist pattern, the anti-reflection
film 18 is formed based on a lift-off method. Thus, the multi-junction solar cell
illustrated in (A) of FIG. 1 is obtained.
[0089] The multi-junction solar cell in Example 3 is configured of the plurality of sub-cells.
By laminating the plurality of sub-cells configured of compound semiconductors having
different band gaps (by achieving multi-junction), solar light that has a wide range
of energy distribution is utilized efficiently. Further, in the multi-junction solar
cell in Example 3, out of the plurality of sub-cells 11, 12, 13, and 14 configured
of compound semiconductor layers having different compositions, at least the sub-cells
having different lattice constants (in Example 3, between the second sub-cell 12 and
the third sub-cell 13, (Lc
1-Lc
2)/Lc
1=3.8×10
-2) are joined to each other with the connection layer 20 in between. The connection
layer 20 is allowed to be formed in a layer-like form as a thin film (for example,
of 5 nm or less), and in particular, achieves ohmic resistance with respect to the
compound semiconductor layer. Also, by using titanium (Ti) having low resistivity,
the contact resistance value of the junction portion is suppressed to 1×10
3 Ω·cm
2 or smaller.
[0090] Usually, in surface activation by plasma application, plasma damage is caused in
the junction surface. However, in Example 3, the connection layers 20A and 20B made
of metal are formed on the surfaces of the second sub-cell 12 and the third sub-cell
13, then, the surfaces of the connection layers 20A and 20B are activated by plasma
application, and thereafter, the connection layers 20A and 20B are joined to each
other. Here, the connection layers 20A and 20B serve as protective films with respect
to the second sub-cell 12 and the third sub-cell 13, and occurrence of plasma damage
in the second sub-cell 12 and the third sub-cell 13 is prevented. Therefore, an increase
in contact resistance caused by the plasma application is prevented. It is to be noted
that the connection layer 20 configured of Ti formed by the vacuum evaporation method
has become a layer having amorphous characteristics because of this plasma application.
It is to be noted that conditions for the plasma application are set to conditions
that allow collision energy of plasma to be relatively weak. In particular, conditions
that cause damage in a region at several-ten nanometers or more from the surface are
not used as usual, and conditions that damage a region only at several nanometers
from the surface are used.
[0091] Moreover, in Example 3, the surfaces of the connection layers 20A and 20B are activated
by the plasma application for junction, and therefore, junction is allowed to be formed
at a low temperature of 150°C or lower. Accordingly, compound semiconductor materials
are allowed to be selected without being limited by thermal expansion coefficients.
In other words, degree of freedom in selecting the compound semiconductor materials
that configure the multi-junction solar cell is increased, and it becomes possible
to select compound semiconductor materials that allow spacings of band gaps to be
equal. Further, occurrence of damage in the junction face due to heating is also prevented.
[0092] By the way, the amounts of an n-type dopant and a p-type dopant added to the respective
compound semiconductor layers are set to allow dopant concentration in the respective
n
+-type and p
+-type compound semiconductor layers to be, for example, about from 1×10
16 cm
-3 to 5×10
19 cm
-3 both inclusive. However, when the dopant concentration of the p
+-GaAs layer is 1×10
19 cm
-3 or higher, light having a long wavelength may not be transmitted because of free
carrier absorption. (A) and (B) of FIG. 15 show results of infrared microscopic transmission
experiments for p
+-GaAs layer (dopant concentration: 2×10
19 cm
-3)/n
+-InP layer (dopant concentration: 4×10
18 cm
-3) and n
+-GaAs layer (dopant concentration: 2×10
18 cm
-3)/n
+-InP layer (dopant concentration: 4×10
18 cm
-3) at a wavelength from 1.1 µm to 1.2 µm both inclusive. It can be seen that, when
the n
+-GaAs layer having low dopant concentration as 2×10
18 cm
-3 is used, light is transmitted as shown in (A) of FIG. 15, and on the other hand,
when the p
+-GaAs layer having high dopant concentration as 2×10
19 cm
-3 is used, light is not transmitted as shown in (B) of FIG. 15. Accordingly, it can
be found that the p
+-GaAs layer having high dopant concentration as 2×10
19 cm
-3 is not transparent with respect to light having a long wavelength. Therefore, when
the film thickness of the p
+-GaAs layer is large, the p
+-GaAs layer becomes an absorption layer, and therefore, it is necessary to reduce
the film thickness thereof depending on design. For example, in the p
+-GaAs layer having dopant concentration as 2×10
19 cm
-3, an absorption coefficient is as large as 2500 cm
-1 with respect to light having photon energy of 0.5 eV (having a wavelength of about
2.5 µm). Therefore, in order to allow light transmittance to be 90% or higher, the
film thickness may be preferably 400 nm or less. Moreover, by allowing the film thickness
to be 40 nm or less, the light transmittance is allowed to be 99% or higher.
[0093] Moreover, in order to improve efficiency in utilizing solar light, it is necessary
to take in solar light spectrum in a wide range. The maximum wavelength of the solar
light spectrum is 2.5 µm. However, as described above, when the concentration of the
p-type dopant is high, light having a long wavelength is difficult to be transmitted.
FIG. 16 shows a relationship between photon energy and absorption coefficients for
each concentration of the p-type dopant in the p-type GaAs layer. It is to be noted
that, in FIG. 16, "A" is data for p-type dopant concentration of 1.5×10
17, "B" is data for p-type dopant concentration of 1.1×10
19, "C" is data for p-type dopant concentration of 2.6×10
19, "D" is data for p-type dopant concentration of 6.0×10
19, and "E" is data for p-type dopant concentration of 1.0×10
20. As can be seen from FIG. 16, the p-type GaAs layer having p-type dopant concentration
of 3×10
19 has an absorption coefficient of 4000 cm
-1 with respect to light having photon energy of 0.5 eV (a wavelength of about 2.5 µm).
FIG. 17 shows a relationship between the thickness of the p-type GaAs layer having
p-type dopant concentration of 3×10
19 and light transmittance of solar light having the maximum wavelength of 2.5 µm based
on the data in FIG. 16. As can be seen from FIG. 17, in order to obtain light transmittance
of solar light of 90% or higher, the film thickness of the p-type GaAs layer may be
set to 270 nm or smaller, and in order to obtain light transmittance of 98% or higher,
the film thickness thereof may be set to 50 nm or smaller. Further, it is found that,
in order to obtain light transmittance of 99% or higher, the film thickness thereof
may be set to 25 nm or smaller.
[0094] For reference, FIG. 18 shows a photograph of a bright-field image of an interface
of the junction of the InP substrate and the GaAs substrate taken by a scanning transmission
electron microscope. Here, an upper part of FIG. 18 is an interface obtained by directly
joining the InP substrate to the GaAs substrate. Further, a middle part and a lower
part of FIG. 18 are interfaces obtained by forming Ti layers having film thicknesses
of 2.3 nm and 1.0 nm, on the InP substrate and the GaAs substrate, respectively, with
the use of an evaporation apparatus of a resistance-heating scheme at vacuum degree
of 2×10
-4 Pa, at evaporation speed of 0.1 nm/sec, at a substrate temperature of 80°C, and at
substrate rotation speed of 30 rpm, and then, joining the two substrates to each other
with these Ti layers in between. As can be seen from the photographs shown in the
middle part and the lower part of FIG. 18, the layer-like Ti layers having substantially
uniform film thicknesses are formed.
[0095] Moreover, for reference, an oxidation state of the Ti layer was examined. Generally,
metal is easier to be naturally-oxidized compared to semiconductors. FIG. 19 shows
variation in light transmittance over time of the Ti layer having a thickness of 2.0
nm for each wavelength. It is to be noted that, in FIG. 19, "A" is data in a case
where the Ti layer is left in atmosphere for 2 hours, "B" is data in a case where
the Ti layer is left in atmosphere for 24 hours, and "C" is data in a case where the
Ti layer is left in atmosphere for 3 months. Further, FIG. 20 shows light transmittance
after 2 hours have elapsed after the film formation (shown as "B" group in FIG. 20),
and light transmittance after 24 hours have elapsed after the film formation (shown
as "A" group in FIG. 20). As can be seen from FIG. 19 and FIG. 20, light transmittance
is increased as the time elapses. In particular, based on FIG. 20, the light transmittance
24 hours after the film formation is increased by from 3% to 6% compared with the
light transmittance 2 hours after the film formation. It can be considered that this
is because a titanium oxide film (TiO
2) is formed on the surface of the Ti film due to exposure to atmosphere, and therefore,
the film thickness of Ti is reduced. When an oxide film such as TiO
2 is formed, contact resistance in the junction interface is increased, and electric
conductivity may be lowered.
[0096] Moreover, for reference, evaluation was done on plasma treatment (plasma application).
The oxide film formed on the surfaces of the connection layers 20A and 20B is removed
by the plasma treatment (the plasma application) at the same time as the time of activation
of the surfaces of the connection layers 20A and 20B. Specifically, energy of ions
that are incident on the surfaces of the connection layers 20A and 20B by the Ar plasma
treatment is utilized to cut the bonding between the metal atom (Ti atom) and an oxygen
atom and to allow the oxygen atom to be separated from the surfaces. FIG. 21 shows
a result of quantitative analysis of concentration of each atom in each distance in
a lamination direction of the multi-junction solar cell based on energy dispersive
X-ray spectrometry (EDX). A content of oxygen (O) in a region at about 10 nm in which
the connection layer 20 is formed is 1/3 or less compared to a content of Ti, and
is sufficiently lower than that of TiO
2 (the number of O atoms is twice of that of the Ti atoms). Accordingly, it can be
found that oxygen is removed by the application of Ar plasma. It is to be noted that,
by the application of Ar plasma, impurities such as Fe, Cr, and Al may be mixed into
the interfaces of the connection layers 20A and 20B from a material of components
configuring a plasma treatment apparatus. However, a particular issue is not caused
in characteristics.
[0097] Evaluation was done on the contact resistance ρ
c of the connection layer. Specifically, in a manner similar to that in [Step-310]
in Example 3, a Ti layer having a thickness of 1.8 nm was formed on the p-type GaAs
substrate. On the other hand, in a manner similar to that in [Step-310] in Example
3, a Ti layer having a thickness of 1.8 nm was formed on the n-type InP substrate.
Further, in a manner similar to that in [Step-320] in Example 3, after performing
plasma treatment on these Ti layers, the Ti layers were joined to each other at an
ambient pressure of 1×10
-4 Pa, at a junction load of 2×10
4 N, and at a temperature of 25°C. Further, an electrode configured of Ti/Pt/Au was
formed on an outer face of the p-type GaAs substrate and an outer face of the n-type
InP substrate. Further, current-voltage characteristics of Sample-1 obtained in such
a manner were measured, and the contact resistance ρ
c in the junction interface was obtained based on the measurement result. Accordingly,
the following result was obtained.

[0098] In Sample-2 in which the thickness of the Ti layer was changed from 1.8 nm to 1.0
nm, the following result was obtained.

[0099] It is to be noted that, in Sample-3 in which an electrode configured of Ti/Pt/Au
was formed on each of the both faces of the p-type GaAs substrate, the following result
was obtained.

[0100] Further, Sample-4 in which an electrode configured of Ti/Pt/Au was formed on each
of the both faces of the n-type InP substrate, the following result was obtained.

[0101] In these measurements, favorable linear ohmic contact was obtained. As can be seen
from the above-described results, ρ
c≤1×10
-3 Ω·cm
2 is achieved when the connection layer 20 is configured of the Ti layer having a thickness
of 5 nm or less. Moreover, the contact resistance of Sample-1 or the contact resistance
of Sample-2 is almost equal to sum of the contact resistance of Sample-3 and the contact
resistance of Sample-4. Accordingly, it is found that electric loss in a case where
the p-type GaAs substrate is joined to the n-type InP substrate with the use of the
connection layer configured of the Ti layer is almost "0", and ideal junction is achieved.
[0102] Moreover, current-voltage characteristics were measured for Sample-5 in which the
surfaces of the p-type GaAs substrate and the n-type InP substrate were allowed to
be in an amorphous state, and the p-type GaAs substrate was joined to the n-type InP
substrate through these surfaces in a method same as that in Sample-1, and for Sample-6
in which the thickness of the Ti layer was changed to 0.5 nm (a fabricating method
is the same as that in Sample-1). As a result, current-voltage characteristics similar
to those of Sample-1 were obtained. Accordingly, it is found that favorable linear
ohmic contact is obtained also in the case where the compound semiconductor layers
are allowed to be in the amorphous state and the junction is formed using them as
the connection layers.
[0103] It is to be noted that, results similar to those described above were obtained also
in a case where the connection layer was configured of Ti layer/Al layer instead of
Ti layer/Ti layer.
EXAMPLE 4
[0104] Example 4 is a modification of Example 3. (A) of FIG. 4 illustrates a conceptual
diagram of a multi-junction solar cell, a photoelectric conversion device, and a compound-semiconductor-layer
lamination structure in Example 4. In Example 4, a connection layer 21 has a lamination
structure configured of a plurality of types (two types, in Example 4) of metal thin
films. Specifically, for example, a Ti layer (a connection layer 21A) having a thickness
of 0.5 nm may be formed on the compound semiconductor layer 12C configured of n
+-In
0.79Ga
0.21As
0.43P
0.57 configuring the second sub-cell 12, and on the other hand, for example, an Al film
(a connection layer 21B) having a thickness of 0.5 nm may be formed on the compound
semiconductor layer 13A configured of p
+-GaAs configuring the third sub-cell 13. Subsequently, Ar plasma application is performed
on these connection layers 21A and 21B to activate surfaces thereof in a manner similar
to that in [Step-320] in Example 3. Also, the connection layers 21A and 21B are allowed
to be amorphous, and then, are joined to each other. FIG. 22 shows a photograph of
a cross-section, of a bonding junction interface, obtained by a transmission electron
microscope. As can be seen from FIG. 22, because the connection layers are amorphous,
crystal lattice is not viewable in the transmission electron microscope image. It
is to be noted that metal used as the connection layer 21 may be selected appropriately
from metal having ohmic characteristics and capable of forming a layer having a thickness
of several nanometers or smaller, that may be, Al, Ti, Zr, Hf, W, Ta, Mo, Nb, or V.
A combination of metal used as the connection layers 21A and 21B is not particularly
limited. Metal exhibiting electric characteristics with favorable ohmic characteristics
with respect to the compound semiconductor layers 12C and 13A forming the sub-cells
12 and 13 may be selected separately for the connection layers 21A and 21 B, respectively.
Further, accordingly, contact resistance is suppressed to the minimum.
EXAMPLE 5
[0105] Example 5 is also a modification of Example 3. Example 5 is different from Example
3 in that a connection layer 22 is configured of amorphous layers of compound semiconductors
configuring the respective second sub-cell 12 and third sub-cell 13. (B) of FIG. 4
illustrates a conceptual diagram of a multi-junction solar cell, a photoelectric conversion
device, and a compound-semiconductor-layer lamination structure in Example 5.
[0106] The connection layer 22 in Example 5 is configured of an n
+-In
0.79Ga
0.21As
0.43P
0.57 amorphous layer (a connection layer 22A) and a p
+-GaAs amorphous layer (a connection layer 22B). The n
+-In
0.79Ga
0.21As
0.43P
0.57 amorphous layer is obtained by allowing part of the compound semiconductor layer
12C configured of n
+-In
0.79Ga
0.21As
0.43P
0.57 configuring the second sub-cell 12 to be amorphous. The p
+-GaAs amorphous layer is obtained by allowing part of the compound semiconductor layer
13A configured of p
+-GaAs configuring the third sub-cell 13 to be amorphous. Dopant concentration of the
connection layer 22A and the connection layer 22B may be, for example, from 1×10
18 cm
-3 to 5×10
19 cm
-3 both inclusive. A film thickness of the connection layer 22 may be preferably from
0.5 nm to 3.0 nm both inclusive, for example. Further, film thicknesses of the connection
layers 22A and 22B may be preferably a half of that of the connection layer 22 after
the junction, that is, from 0.25 nm to 1.5 nm both inclusive.
[0107] In Example 5, after forming the compound semiconductor layers, in a manner similar
to that in [Step-320] in Example 3, the surfaces of the compound semiconductor layer
12C and the compound semiconductor layer 13A are activated by plasma treatment, and
also, after the surfaces are allowed to be amorphous, the second sub-cell 12 is joined
to the third sub-cell 13. Specifically, Ar plasma (for example, at plasma density
from 10
9 cm
-3 to 10
11 cm
-3 both inclusive, and at a pressure from 1 Pa to 10
-2 Pa both inclusive) is applied to surfaces of the compound semiconductor layer 12C
configured of n
+-In
0.79Ga
0.21As
0.43P
0.57 and the compound semiconductor layer 13A configured of p
+-GaAs, and plasma damage is caused on the surface of each of the compound semiconductor
layers 12C and 13A. Accordingly, for example, an amorphous layer (the connection layers
22A and 22B) having a film thickness of 1.0 nm may be formed. Further, the connection
layers 22A and 22B are bonded to each other while maintaining high degree of vacuum
(for example, at 5×10
-4 Pa or lower), at junction load of 2×10
4 N or less and at a temperature of 150°C or lower, in particular, for example, at
an ambient pressure of 1×10
-4 Pa, at junction load of 2×10
4 N, and at a temperature of 25°C, and thereby, the second sub-cell 12 is joined to
the third sub-cell 13.
[0108] In Example 5, the crystal structure of part of the compound semiconductor layers
configuring each sub-cell is allowed to be amorphous between the sub-cells having
different lattice constants, and this is used as the connection layers 22A and 22B.
Accordingly, in a manner similar to that in Example 3, a multi-junction solar cell
in which contact resistance in the junction interface of the joined compound semiconductor
layers is low and high energy conversion efficiency is achieved is obtained. Further,
in addition to such an effect, the step of forming the connection layer configured
of metal becomes unnecessary. Therefore, manufacturing process is simplified, and
manufacture cost is reduced.
EXAMPLE 6
[0109] Example 6 is also a modification of Example 3. Example 6 is different from Example
3 in that a first substrate for film formation and a second substrate for film formation
are used, and these first substrate for film formation and second substrate for film
formation are peeled off at last.
[0110] Description will be given below of a method of manufacturing a multi-junction solar
cell and a photoelectric conversion device in Example 6 referring to (A) to (B) of
FIG. 5, (A) to (B) of FIG. 6, and FIG. 7 that are conceptual diagrams of the compound
semiconductor layers and the like.
[Step-600]
[0111] First, a first sacrificial layer for peeling-off 42 made of AlInAs and an n
+-InP layer 43 that is to serve as a contact layer are formed on a first substrate
for film formation 41 configured of an n-type InP substrate, and then, the second
sub-cell 12, the first tunnel junction layer 15, and the first sub-cell 11 are sequentially
formed on the n
+-InP layer 43. However, the formation of the n
+-InP layer 43 is not essential, and the formation may be omitted as in Example 3 to
Example 5. The same is applicable also to Example 7 which will be described later.
On the other hand, a second sacrificial layer for peeling-off 46 made of AlAs is formed
on a second substrate for film formation 44 made of an n-type GaAs substrate, and
then, the window layer 17, the fourth sub-cell 14, the second tunnel junction layer
16, and the third sub-cell 13 are sequentially formed. Thus, a structure shown in
the conceptual diagram in (A) of FIG. 5 is obtained. It is to be noted that the n
+-InP layer 43 that is to serve as the contact layer may be formed in the multi-junction
solar cells described in Example 3 to Example 5.
[Step-610]
[0112] Further, the first substrate for film formation 41 is peeled off by removing the
first sacrificial layer for peeling-off 42 by etching after the surface of the first
sub-cell 11 is bonded to the support substrate 32 (see (B) of FIG. 5). Thereafter,
for example, the connection layer 20A made of Ti may be formed on the n
+-InP layer 43 formed on the compound semiconductor layer 12C made of n
+-In
0.79Ga
0.21As
0.43P
0.57 configuring the second sub-cell 12. On the other hand, for example, the connection
layer 20B made of Ti may be formed on the compound semiconductor layer 13A made of
p
+-GaAs configuring the third sub-cell 13. It is to be noted that the connection layers
20A and 20B are allowed to be formed in a manner similar to that in [Step-310] in
Example 3. Thus, a structure shown in the conceptual diagram in (A) of FIG. 6 is obtained.
[Step-620]
[0113] Next, in a manner similar to that in [Step-320] in Example 3, the surfaces of the
connection layers 20A and 20B are activated by Ar plasma application, and at the same
time, are allowed to be amorphous. Thereafter, the connection layers 20A and 20B are
joined to each other (see (B) of FIG. 6). Thereafter, after the second substrate for
film formation 44 is peeled off by removing the second sacrificial layer for peeling-off
46 by etching, the second electrode 19 and the anti-reflection film 18 are formed
in a manner similar to that in [Step-330] in Example 3. Thus, the multi-junction solar
cell in Example 6 shown in the conceptual diagram in FIG. 7 is obtained.
[0114] In Example 6, not only the second substrate for film formation but also the first
substrate for film formation is peeled off. Accordingly, the n-type GaAs substrate
and the n-type InP substrate are both allowed to be reused, and therefore, manufacturing
cost is further reduced.
[0115] It is to be noted that, in Example 6, although the connection layer is configured
of Ti as in Example 3, the connection layer may have a configuration similar to that
in Example 4 or Example 5. The same is applicable also to Example 7 which will be
described next.
EXAMPLE 7
[0116] Example 7 is a modification of Example 6. Example 7 is different from Example 6 in
that the first substrate for film formation and the second substrate for film formation
are peeled off after the second sub-cell and the first sub-cell are formed on the
first substrate for film formation and the third sub-cell and the fourth sub-cell
are formed on the second substrate for film formation.
[0117] Description will be given below of a method of manufacturing a multi-junction solar
cell and a photoelectric conversion device in Example 7 referring to (A) to (B) of
FIG. 8, (A) to (B) of FIG. 9, and (A) to (B) of FIG. 10 that are conceptual diagrams
of the compound semiconductor layers and the like.
[Step-700]
[0118] First, in a manner similar to that in [Step-600] in Example 6, the first sacrificial
layer for peeling-off 42, the n
+-InP layer 43, the second sub-cell 12, the first tunnel junction layer 15, and the
first sub-cell 11 are sequentially formed on the first substrate for film formation
41 configured of the n-type InP substrate. On the other hand, the second sacrificial
layer for peeling-off 46, the third sub-cell 13, the second tunnel junction layer
16, the fourth sub-cell 14, the window layer 17, and a third sacrificial layer for
peeling-off 47 are sequentially formed on the second substrate for film formation
44 configured of an n-type GaAs substrate. Thus, a structure shown in the conceptual
diagram in (A) of FIG. 8 is obtained.
[Step-710]
[0119] Thereafter, the first substrate for film formation 41 is peeled off by removing the
first sacrificial layer for peeling-off 42 by etching. Further, the second substrate
for film formation 44 is peeled off by removing the second sacrificial layer for peeling-off
46 by etching. Thus, a structure shown in the conceptual diagram in (B) of FIG. 8
is obtained.
[Step-720]
[0120] Next, for example, the connection layer 20A made of Ti may be formed on the n
+-InP layer 43 formed on the compound semiconductor layer 12C made of n
+-In
0.79Ga
0.21As
0.43P
0.57 configuring the second sub-cell 12. On the other hand, for example, the connection
layer 20B made of Ti may be formed on the compound semiconductor layer 13A made of
p
+-GaAs configuring the third sub-cell 13. It is to be noted that the connection layers
20A and 20B are allowed to be formed in a manner similar to that in [Step-310] in
Example 3. Thus, a structure shown in the conceptual diagram in (A) of FIG. 9 is obtained.
[Step-730]
[0121] Thereafter, for example, the first sub-cell 11 may be bonded to the support substrate
33 and the third sacrificial layer for peeling-off 47 may be bonded to the support
substrate 34 with the use of wax, a resist having high viscosity, or the like. Thus,
a structure shown in the conceptual diagram in (B) of FIG. 9 is obtained.
[Step-740]
[0122] Subsequently, in a manner similar to that in [Step-320] in Example 3, the surfaces
of the connection layers 20A and 20B are activated by Ar plasma application, and at
the same time, are allowed to be amorphous. Thereafter, the connection layers 20A
and 20B are joined to each other (see (A) of FIG. 10). Thereafter, the support substrate
34 is peeled off by removing the third sacrificial layer for peeling-off 47 by etching.
Subsequently, the second electrode 19 and the anti-reflection film 18 are formed in
a manner similar to that in [Step-330] in Example 3. Thus, the multi-junction solar
cell in Example 7 shown in the conceptual diagram in (B) of FIG. 10 is obtained.
[0123] Hereinabove, the present disclosure has been described based on preferable Examples.
However, the present disclosure is not limited to these Examples. The configurations,
the structures, the compositions, and the like of the multi-junction solar cell, the
photoelectric conversion device, the compound-semiconductor-layer lamination structure,
or the like in the Examples are allowed to be changed as appropriate. It is not necessary
to provide all of the various compound semiconductor layers configuring the multi-junction
solar cells, the photoelectric conversion devices, and the like described in the Examples,
and other layers may be provided. Further, the junction of the connection layers 20A
and 20B may be formed, for example, at 200°C, and thereby, contact resistance of the
junction interface is further reduced.
[0124] Since the substrate for film formation is removed at last, the conductivity type
of the substrate may be either the n-type or the p-type. Further, since the substrate
for film formation is allowed to be reused, the manufacturing cost of the multi-junction
solar cell, the photoelectric conversion device, and the like is reduced.
[0125] For example, the multi-junction solar cell shown in the conceptual diagram in (A)
of FIG. 1 may be a structure in which the connection layer 20 is extended to an outer
side and may configure a third electrode as shown in FIG. 11. Accordingly, a parallel
multi-junction solar cell is configured that is allowed to easily face a region in
which spectrum of solar light is different from AM1.5, variation in climate, etc.
It is to be noted that, in the multi-junction solar cell shown in FIG. 11, a lamination
order of compound semiconductor layers 11A
1, 11A
2, and 11C, and the compound semiconductor layers 12A, 12B, and 12C in the first sub-cell
11 and the second sub-cell 12 is the reverse of a lamination order of the compound
semiconductor layers 11A
1, 11A
2, and 11C and the compound semiconductor layers 12A, 12B, and 12C in the first sub-cell
11 and the second sub-cell 12 in the multi-junction solar cell in Example 1 and Example
3 shown in (A) of FIG. 1, and the first sub-cell 11 and the second sub-cell 12 are
connected in parallel to the third sub-cell 13 and the fourth sub-cell 14. Further,
in order to reduce electric resistance to the third electrode, for example, a thickness
of the p
+-InP layer 43 may be desirably larger.
[0126] In the Examples, a configuration including, in order from the light incident side,
for example,
Fourth sub-cell: InGaP layer
Third sub-cell: GaAs layer
Second sub-cell: InGanAsP layer
First sub-cell: InGaAs layer
has been adopted. However, alternatively, for example, [Configuration-A] to [Configuration-D]
that are configurations shown in Table 6 below in order from the light incident side
may be adopted. Alternatively, Table 7 below shows [Configuration-E] to [Configuration-G]
that are each a configuration in which the third sub-cell, the fourth sub-cell, and
the fifth sub-cell are formed on the GaAs substrate, the first sub-cell and the second
sub-cell are formed on the InP substrate, and the second sub-cell is joined to the
third sub-cell. Moreover,
[0127] Table 8 below shows [Configuration-H] to [Configuration-K] that are each a configuration
in which the second sub-cell, the third sub-cell, and the fourth sub-cell are formed
on the GaAs substrate, the first sub-cell is formed on the InP substrate, and the
first sub-cell is joined to the second sub-cell. The first sub-cell in Table 6 to
Table 8 corresponds to the first sub-cell in the Examples, and has the distortion
compensation lamination structure. It is to be noted that a third column in Table
6 to Table 8 shows values of band gaps, and a fourth column therein shows values of
lattice constants. Further, the value of the lattice constant of the first sub-cell
is an average lattice constant value. Further, in Table 6 to Table 8, compound semiconductor
layers that have the same composition but have different values of band gaps or different
values of lattice constants have different atomic percentages.
[Table 6]
[Configuration-A] |
Fourth sub-cell |
InGaP layer |
1.90 eV |
5.653 Å |
Third sub-cell |
In0.01 Ga0.99As layer |
1.40 eV |
5.657 Å |
Second sub-cell |
InGaAsP laver |
1.03 eV |
5.868 Å |
First sub-cell |
In0.32Ga0.68As/ In0.74Ga0.26As layer |
0.55 eV |
5.868 Å |
[Configuration-B] |
Fourth sub-cell |
AlInGaP laver |
2.10 eV |
5.722 Å |
Third sub-cell |
In0.65Ga0.35P layer |
1.66 eV |
5.722 Å |
Second sub-cell |
InGaAsP layer |
1.03 eV |
5.868 Å |
First sub-cell |
In0.32Ga0.68As/ In0.74Ga0.26As layer |
0.55 eV |
5.868 Å |
|
[Configuration-C] |
Fourth sub-cell |
InGaP laver |
1.90 eV |
5.653 Å |
Third sub-cell |
In0.01Ga0.99As layer |
1.40 eV |
5.657 Å |
Second sub-cell |
InGaAsP laver |
1.03 eV |
5.868 Å |
First sub-cell |
In0.45Ga0.55As/ In0.63Ga0.37As laver |
0.65 eV |
5.868 Å |
|
[Configuration-D] |
Fourth sub-cell |
AlInGaP laver |
2.10 eV |
5.722 Å |
Third sub-cell |
In0.65Ga0.35P layer |
1.66 eV |
5.722 Å |
Second sub-cell |
InGaAsP laver |
1.03 eV |
5.868 Å |
First sub-cell |
In0.45Ga0.55As/ In0.63Ga0.37As laver |
0.65 eV |
5.868 Å |
[Table 7]
[Configuration-E] |
Fifth sub-cell |
Al0.015InGa0.985P layer |
1.90 eV |
5.653 Å |
Fourth sub-cell |
GaAs laver |
1.42 eV |
5.653 Å |
Third sub-cell |
InGaAsN layer |
1.00 eV |
5.653 Å |
Second sub-cell |
In0.53Ga0.47As laver |
0.75 eV |
5.868 Å |
First sub-cell |
In0.32Ga0.68As/ In0.74Ga0.26As layer |
0.55 eV |
5.868 Å |
[Configuration-F] |
Fifth sub-cell |
Al0.015InGa0.985P layer |
1.90 eV |
5.653 Å |
Fourth sub-cell |
In0.01Ga0.99As layer |
1.40 eV |
5.657 Å |
Third sub-cell |
InGaAsN layer |
1.00 eV |
5.653 Å |
Second sub-cell |
In0.53Ga0.47As laver |
0.75 eV |
5.868 Å |
First sub-cell |
In0.32Ga0.68As/ In0.74Ga0.26As layer |
0.55 eV |
5.868 Å |
|
[Configuration-G] |
Fifth sub-cell |
Al0.27InGa0.73P laver |
2.20 eV |
5.653 Å |
Fourth sub-cell |
InGaP laver |
1.80 eV |
5.657 Å |
Third sub-cell |
GaAs laver |
1.42 eV |
5.653 Å |
Second sub-cell |
InGaAsP laver |
1.03 eV |
5.868 Å |
First sub-cell |
In0.45Ga0.55As/ In0.63Ga0.37As layer |
0.65 eV |
5.868 Å |
[Table 8]
[Configuration-H] |
Fourth sub-cell |
InGaP laver |
1.90 eV |
5.653 Å |
Third sub-cell |
In0.01Ga0.99As layer |
1.40 eV |
5.657 Å |
Second sub-cell |
InGaAsN layer |
1.00 eV |
5.653 Å |
First sub-cell |
In0.32Ga0.68As/ In0.74Ga0.26As layer |
0.55 eV |
5.868 Å |
|
[Configuration-I] |
Fourth sub-cell |
InGaP layer |
1.90 eV |
5.653 Å |
Third sub-cell |
GaAs layer |
1.42 eV |
5.653 Å |
Second sub-cell |
InGaAsN laver |
1.00 eV |
5.653 Å |
First sub-cell |
In0.32Ga0.68As/ In0.74Ga0.26As laver |
0.55 eV |
5.868 Å |
[Configuration-J] |
Fourth sub-cell |
InGaP laver |
1.90 eV |
5.653 Å |
Third sub-cell |
In0.01Ga0.99As layer |
1.40 eV |
5.657 Å |
Second sub-cell |
InGaAsN layer |
1.00 eV |
5.653 Å |
First sub-cell |
In0.45Ga0.55As/ In0.63Ga0.37As layer |
0.65 eV |
5.868 Å |
|
[Configuration-K] |
Fourth sub-cell |
InGaP layer |
1.90 eV |
5.653 Å |
Third sub-cell |
GaAs layer |
1.42 eV |
5.653 Å |
Second sub-cell |
InGaAsN layer |
1.00 eV |
5.653 Å |
First sub-cell |
In0.45Ga0.55As/ In0.63Ga0.37As layer |
0.65 eV |
5.868 Å |
[0128] Moreover, it is not limited to the four-junction type as described above, and a multi-junction
solar cell having less-than-four junctions may be achieved, or a multi-junction solar
cell having five-or-more junctions (for example, AlInGaP layer/InGaP layer/AlGaAs
layer/ InGaAs layer/ InGaAsN layer/ Ge layer) may be achieved.
[0129] It is to be noted that the present disclosure may also have configurations such as
the followings.
- [1] [Solar cell]
A multi-junction solar cell,
the multi-junction solar cell including a plurality of sub-cells that are laminated,
in which light enters from the sub-cell located in an uppermost layer to the sub-cell
located in a lowermost layer, and electric power is generated in each of the sub-cells,
each of the sub-cells including a first compound semiconductor layer and a second
compound semiconductor layer that are laminated, the first compound semiconductor
layer having a first conductivity type, and the second compound semiconductor layer
having a second conductivity type, wherein,
in at least one predetermined sub-cell of the plurality of sub-cells,
the first compound semiconductor layer is configured of at least one first-compound-semiconductor-layer
lamination unit in which a 1-A compound semiconductor layer and a 1-B compound semiconductor
layer are laminated, the 1-A compound semiconductor layer having the first conductivity
type, and the 1-B compound semiconductor layer having the first conductivity type
the second conductivity type,
the second compound semiconductor layer is configured of at least one second-compound-semiconductor-layer
lamination unit in which a 2-A compound semiconductor layer and a 2-B compound semiconductor
layer are laminated, the 2-A compound semiconductor layer having the second conductivity
type that is different from the first conductivity type, and the 2-B compound semiconductor
layer having the second conductivity type,
a compound semiconductor composition configuring the 1-A compound semiconductor layer
and a compound semiconductor composition configuring the 2-A compound semiconductor
layer are same compound semiconductor composition A,
a compound semiconductor composition configuring the 1-B compound semiconductor layer
and a compound semiconductor composition configuring the 2-B compound semiconductor
layer are same compound semiconductor composition B,
the compound semiconductor composition A is determined based on a value of a band
gap of the predetermined sub-cell,
the compound semiconductor composition B is determined based on a difference between
a base lattice constant of a base used at a time of forming the first and second compound
semiconductor layers and a lattice constant of the compound semiconductor composition
A,
a thickness of the 1-B compound semiconductor layer is determined based on a difference
between the base lattice constant and a lattice constant of the compound semiconductor
composition B, and on a thickness of the 1-A compound semiconductor layer,
a thickness of the 2-B compound semiconductor layer is determined based on the difference
between the base lattice constant and the lattice constant of the compound semiconductor
composition B, and on a thickness of the 2-A compound semiconductor layer,
the thicknesses of the 1-A compound semiconductor layer and the 2-A compound semiconductor
layer are smaller than a critical film thickness of the compound semiconductor composition
A, and are thicknesses that cause no quantum effect, and
the thicknesses of the 1-B compound semiconductor layer and the 2-B compound semiconductor
layer are smaller than a critical film thickness of the compound semiconductor composition
B, and are thicknesses that cause no quantum effect.
- [2] The multi-junction solar cell according to [1], wherein the predetermined sub-cell
is located in the lowermost layer.
- [3] The multi-junction solar cell according to [1] or [2], wherein a group of atoms
configuring the compound semiconductor composition A is same as a group of atoms configuring
the compound semiconductor composition B.
- [4] The multi-junction solar cell according to [3], wherein atomic percentage of the
group of atoms configuring the compound semiconductor composition A is different from
atomic percentage of the group of atoms configuring the compound semiconductor composition
B.
- [5] The multi-junction solar cell according to any one of [1] to [4], wherein the
value of the band gap of the predetermined sub-cell is from 0.45 eV to 0.75 eV both
inclusive.
- [6] The multi-junction solar cell according to any one of [1] to [5], wherein a value
of a band gap of the compound semiconductor composition B is larger than a value of
a band gap of the compound semiconductor composition A.
- [7] The multi-junction solar cell according to any one of [1] to [6], wherein

and

are satisfied where LCA is the lattice constant of the compound semiconductor composition A, LCB is the lattice constant of the compound semiconductor composition B, and LC0 is the base lattice constant.
- [8] The multi-junction solar cell according to [7], wherein

and

are established.
- [9] The multi-junction solar cell according to any one of [1] to [8], wherein

is satisfied where LCA is the lattice constant of the compound semiconductor composition A, LCB is the lattice constant of the compound semiconductor composition B, LC0 is the base lattice constant, tA is the thickness of the 1-A compound semiconductor layer and the 2-A compound semiconductor
layer, and tB is the thickness of the 1-B compound semiconductor layer and the 2-B compound semiconductor
layer.
- [10] The multi-junction solar cell according to any one of [1] to [9], wherein the
base is formed of InP, the compound semiconductor composition A is InxGa1-xAs, and the compound semiconductor composition B is InyGa1-yAs (where x>y).
- [11] The multi-junction solar cell according to [10], wherein 0.53≤x≤0.86 and 0≤y≤0.53
are established.
- [12] [Photoelectric conversion device]
A photoelectric conversion device,
the photoelectric conversion device including a first compound semiconductor layer
and a second compound semiconductor layer that are laminated, the first compound semiconductor
layer having a first conductivity type, and the second compound semiconductor layer
having a second conductivity type, wherein
the first compound semiconductor layer is configured of at least one first-compound-semiconductor-layer
lamination unit in which a 1-A compound semiconductor layer and a 1-B compound semiconductor
layer are laminated, the 1-A compound semiconductor layer having the first conductivity
type, and the 1-B compound semiconductor layer having the first conductivity type
the second conductivity type,
the second compound semiconductor layer is configured of at least one second-compound-semiconductor-layer
lamination unit in which a 2-A compound semiconductor layer and a 2-B compound semiconductor
layer are laminated, the 2-A compound semiconductor layer having the second conductivity
type that is different from the first conductivity type, and the 2-B compound semiconductor
layer having the second conductivity type,
a compound semiconductor composition configuring the 1-A compound semiconductor layer
and a compound semiconductor composition configuring the 2-A compound semiconductor
layer are same compound semiconductor composition A,
a compound semiconductor composition configuring the 1-B compound semiconductor layer
and a compound semiconductor composition configuring the 2-B compound semiconductor
layer are same compound semiconductor composition B,
the compound semiconductor composition A is determined based on a value of a band
gap of the photoelectric conversion device,
the compound semiconductor composition B is determined based on a difference between
a base lattice constant of a base used at a time of forming the first and second compound
semiconductor layers and a lattice constant of the compound semiconductor composition
A,
a thickness of the 1-B compound semiconductor layer is determined based on a difference
between the base lattice constant and a lattice constant of the compound semiconductor
composition B, and on a thickness of the 1-A compound semiconductor layer,
a thickness of the 2-B compound semiconductor layer is determined based on the difference
between the base lattice constant and the lattice constant of the compound semiconductor
composition B, and on a thickness of the 2-A compound semiconductor layer,
the thicknesses of the 1-A compound semiconductor layer and the 2-A compound semiconductor
layer are smaller than a critical film thickness of the compound semiconductor composition
A, and are thicknesses that cause no quantum effect, and
the thicknesses of the 1-B compound semiconductor layer and the 2-B compound semiconductor
layer are smaller than a critical film thickness of the compound semiconductor composition
B, and are thicknesses that cause no quantum effect.
- [13] [Compound-semiconductor-layer lamination structure]
A compound-semiconductor-layer lamination structure,
the compound-semiconductor-layer lamination structure including at least one compound-semiconductor-layer
lamination unit in which an A compound semiconductor layer and a B compound semiconductor
layer are laminated,
a compound semiconductor composition B configuring the B compound semiconductor layer
is determined based on a difference between a base lattice constant of a base used
at a time of forming the A and B compound semiconductor layers and a lattice constant
of a compound semiconductor composition A configuring the A compound semiconductor
layer,
a thickness of the B compound semiconductor layer is determined based on a difference
between the base lattice constant and a lattice constant of the compound semiconductor
composition B, and on a thickness of the A compound semiconductor layer,
the thickness of the A compound semiconductor layer is smaller than a critical film
thickness of the compound semiconductor composition A, and is a thickness that causes
no quantum effect, and
the thicknesses of the B compound semiconductor layer is smaller than a critical film
thickness of the compound semiconductor composition B, and is a thickness that causes
no quantum effect.
[0130] This application claims the priority on the basis of P.R.C Patent Application No.
201110281329.6 filed September 21, 2011 in State Intellectual Property Office of The P.R.C, the entire contents of each which
are incorporated in this application by reference.